Skip to main content

Full text of "Personality"

See other formats













To my Mother 


As a rule, science regards the individual as a mere bothersome ac- 
cident. Psychology, too, ordinarily treats him as something to be 
brushed aside so the main business of accounting for the uniformity 
of events can get under way. The result is that on all sides we see 
psychologists enthusiastically at work upon a somewhat shadowy 
portrait entitled “the generalized human mind.” Though serving well 
a certain purpose, this portrait is not altogether satisfying to those 
who compare it with the living individual models from which it is 
drawn. It seems unreal and esoteric, devoid of locus, self-conscious- 
ness, and organic unity— all essential characteristics of the minds we 

With the intention of supplementing this abstract portrait by one 
that is more life-like, a new movement within psychological science 
has gradually grown up. It attempts in a variety of ways and from 
many points of view to depict and account for the manifest individ- 
uality of mind. This new movement has come to be known (in 
America) as the psychology of personality. Especially within the past 
fifteen years has its progress been notable. 

Since it is young, this movement finds difficulty in evaluating its 
first achievements. Its research is plentiful but piecemeal; its theories 
are numerous but conflicting. Yet every year more and more psycho- 
logical investigators are attracted to it, and colleges at a rapid rate are 
adding the study of personality to their psychological curricula. The 
result of this rising tide of interest is an insistent demand for a guide 
book that will define the new field of study— one that will articulate 
its objectives, formulate its standards, and test the progress made 
thus far. 

In attempting to write such a book I have sought above all else to 
respect the many-sidedness of the subject-matter of this new science. 
An account written exclusively in terms of any single school of psy- 
chological doctrine would be far too narrow. Better to expand and re- 
fashion one’s theories until they do some measure of justice to the 
richness and dignity of human personality, than to clip and compress 
personality until it fits one closed system of thought. 



In striving for adequacy and balance I have tried to make a special 
ally of common sense which, I believe, affords precisely the hypoth- 
eses and insights that it is the duty of the new psychology of per- 
sonality to verify and (if possible) improve. I have likewise bor- 
rowed liberally from many types of psychological writing, past and 
present. But whatever I have appropriated I have tried to assimilate 
within a single and coherent theoretical frame. 

This goal of adequacy means, of course, that I cannot accept 
whole-heartedly each and every partisan point of view. The endo- 
crinological approach, for example, is a specialty with many enthusi- 
astic supporters. So too is psychoanalysis. Their danger is their one- 
sidedness, their monosymptomatic bias. Though I borrow from these 
approaches I cannot subscribe to them as adequate. 

Similar is the case of the currently popular statistical methodologies. 
Many believe these are indispensable in supplying the factual ground 
for the science of personality. Sometimes they are useful; but many 
times they are not. In any event, mere arrays of statistics are never 
capable of self-interpretation. It is for this reason that I have preferred 
in most cases to state the results of research as clearly as possible in 
words, proceeding at once with the interpretation of the results. If 
the argument is sound, statistics can do no more than symbolize the 
fact; if the argument is unsound, statistical elaboration can never make 
it sound and may even increase the confusion. So, at a time when in 
many quarters mathematical symbolizing enjoys exaggerated favor, I 
prefer for clarity’s sake to stick to the verbal method of exposition 
and argument, especially since it seems to me the only one wherewith 
to co-ordinate the field as a whole. 

From another direction I may be called to task for overlooking 
the close relationship between personality and culture. But such criti- 
cism can arise only from a misunderstanding of my purpose. I do not 
deny that personality is fashioned to a large extent through the impact 
of culture upon the individual. But the interest of psychology is not 
in the factors shaping personality, rather in personality itself as a 
developing structure. From this point of view culture is relevant only 
when it has become interiorized within the person as a set of personal 
ideals, attitudes, and traits. Likewise, culture conflict must become 
inner conflict before it can have any significance for personality. 
Why is it that in our times, when Western culture is sadly disorgan- 
ized, our personalities are not correspondingly disorganized? The 
enthusiastic determinist might reply: “They are. Our institutional 




anchors are lost and each of us is either drifting or breaking to 
pieces.” But such a reply would be wholly unrealistic. Are personali- 
ties in fact any more disorganized now than formerly? Is there any 
sure evidence for an increase of insanity? It is doubtful. Certainly, 
it is impossible to hold that disorganization of personality today is 
proportional to the rapid shattering of cultural forms. Cultural deter- 
minism is one of the monosymptomatic approaches; it has a blind spot 
for the internal balancing factors and structural tenacity within per- 

There is also in some departments of social science a tendency to 
define personality as one man’s influence upon others, as his status in 
the group, or as his “social stimulus-value.” With such definitions 
psychology cannot possibly operate. If it tried to do so its datum 
would evaporate, and there would be left only the notoriously con- 
flicting images that men have of one another. The psychology of 
personality must regard its subject-matter as wholly objective and 
accessible. To be sure, the task of judging personality correctly, of 
reading motives aright, and of representing adequately the change 
and variation of which each person is capable, complicate the study 
enormously, but still a stable biophysical frame of reference must be 

Psychologically considered the important fact about personality is 
its relatively enduring and unique organization. The central problem 
of the psychology of personality therefore concerns the nature of this 
structure and its composition in terms of sub-structures or units. The 
elements and bonds sponsored by traditional psychology do not serve 
as adequate means for depicting the structure of individuality. Part III 
devotes itself entirely to this question, and it is here that the chief 
novelty of my own position lies. Chapters x i and 1 2 on traits , espe- 
cially if taken in conjunction with Chapter 7 on the autonomy of 
motives, supply, I believe, a theory that is concretely applicable to 
the infinitely varied forms of personal existence, and at the same time 
abstract enough to serve as a unifying principle for the new branch 
of science. 

To sum up, my purpose is twofold: (1) to gather into a single 
comprehensive survey the most important fruits of the psychological 
study of personality, and (2) to supply new co-ordinating concepts 
and theories where they will equip this new department of psychol- 
ogy for a more adequate handling of its endlessly rich subject-matter. 

The beginnings of this book lay in certain researches I undertook 



seventeen years ago. Ever since that time it has been in the process of 
development and completion. From start to finish my constant and 
loyal collaborator has been Ada L. Allport, my wife. The material 
has been presented many times in my classes. Through their interest, 
discussion, and willing participation in experiments my students have 
contributed to its final content and form more than they know. In 
certain chapters I have benefited much from the advice and assistance 
of my friends, H. D. Spoerl, C. E. MacGill, D. M. McGregor, R. P. 
Casey, C. M. Harsh, and H. Werner. Especially deep is my indebted- 
ness to my brother, F. H. Allport, for significant help with some of 
the crucial portions of the argument, and to Hadley Cantril who has 
carefully read and criticized the entire manuscript. I wish also to 
acknowledge the kind assistance of R. T. Fuller in drawing the illus- 
trations, and of Miss Dorothy Telfer in preparing the manuscript for 

August, 1937. 

G. W. A. 



Part 1: The Approach to Personality 



Science and the Single Case — Approaches to the Individual within 
the Science of Psychology — Extending the Horizons of Psy- 
chology ......... 



Etymology and Early History of Persona — Theological Meanings — 
Philosophical Meanings — juristic Meanings — Sociological Mean- 
ings — External Appearance (Biosocial Definitions) — Psychologi- 
cal Meanings — A Definition for This Book — Summary — Char- 
acter— Temperament ........ 



Literary Characterology — Humoral Psychology — Physiognomy — 
Phrenology — Ethology and the Study of Sentiments — The Be- 
ginning of Experimental Characterology — Summary . 

Part II: The Development of Personality * 



Heredity — The Beginnings of Personality — Motivation — The Bio- 
logical Theory of Personality — Personality in the First Year 


Differentiation — Integration — Maturation — “Learning’’ 




2 4 








Consciousness of Self — Suggestion -Self-Esteem — Feelings of In- 
feriority and Compensation — Psychoanalytic “Mechanisms” . 159 



Functional Autonomy — Evidence for Functional Autonomy — Cri- 
tique of Functional Autonomy — Sudden Reorientation: Trauma 190 



Extension of the Self -Self-Objectification: Insight and Humor — 

The Unifying Philosophy of Life . . . . . .213 

Part III: The Structure of Personality 



Practical and A Priori Classifications — Uniform (Nomothetic) Ele- 
ments — Stimulus-Response Elements (Specificity) — Summary of 
Arguments Against Specificity — Conclusion . . . . 235 



Partial Identity and Learning — Experimental Work — How Ele- 
mentary is an Element? — How Identical is an Identity? — The 
Question of Proportionality — So-called Identities of Procedure — 
Generalization — Equivalence and Similarity — Summary . . 259 



Are Traits Biosocial or Biophysical? — Traits and Determining 
Tendencies —Trait and Type — Individual versus Common Traits 
— The Problem of Trait-Names 286 






How are Traits Discovered? —The Dynamic Nature of Traits - 
Genotypical, Phenotypical, and Pseudo-Traits — The Independ- 
ence of Traits — The Consistency of Traits — Are Traits Nor- 
mally Distributed? — Cardinal, Central, and Secondary Traits — 
Traits and the Total Personality — Resume of the Doctrine of 
Traits . . . . . . . . . . .312 



Retrospect — Unity as a Philosophical Principle — Unity as Striving — 
Empirical Studies of Unity — Congruence and Methods for Its 
Discovery — A Gestalt Representation of Unity . . . 343 

Part IV: The Analysis of Personality 



Studies of Cultural Setting — Physical Records — Social Records — 
Personal Records — Expressive Movement — Rating — Standardized 
T ests — Statistical Analysis — Miniature Life-Situations — Labora- 
tory Experiments — Prediction — Depth- Analysis — Ideal Types — 
Synthetic Methods — Suggestions for the Preparation of a Case 
Study — Generalization of Case Studies — Conclusions . . 369 

chapter xv 


Construction of the Psychograph — Psycho-Biological Factors — 
Expressive Traits — Attitudinai Traits — The Intercorrelation of 
Traits — Common Traits Not Included in the Psychograph . . 400 



Principles of Rating — Principles of Testing — Principles of Experi- 
mentation . . . . 435 



Expressive versus Adaptive Behavior — Genesis of Expressive Be- 
havior — Psychodiagnostics — The Consistency of Expression — Ex- 
pressive Features — Style — Summary ..... 464 



Part V: Understanding Personality 




First Impressions — Three Basic Factors Involved in Judging Per- 
sonality — The Interview — Is the Ability to Judge Personality 
General or Specific? — Qualifications for a Good Judge of Others 
— Sex Differences in Ability to Judge People— Whom Does One 
Know Best? — Common Sources of Error ..... 499 



Inference — Criticisms of Inference — Empathy — Intuition — The 
Psychology of V erstehen — Criticisms of Intuition — The Place of 
Intuition in the Study of Personality — The Empirical-Intuitive 
Nature of Understanding . 523 



Personalistic Psychology — The Psychology of Personality — Per- 

spective .......... 549 


. 583 


Part I 




Die Natur scheint Alles auf Individualist angelegt zu haben. 


The outstanding characteristic of man is his individuality. He is a 
unique creation of the forces of nature. Separated spatially from all 
other men he behaves throughout his own particular span of life in 
his own distinctive fashion. It is not upon the cell nor upon the single 
organ, nor upon the group, nor upon the species that nature has 
centered her most lavish concern, but rather upon the integral organ- 
ization of life processes into the amazingly stable and self-contained 
system of the individual living creature. 

In daily life, in our direct contacts with our fellows, the pre-emi- 
nence of individuality is recognized readily enough. During our 
waking hours and in our dreams people appear to us as definite and 
individual. The man in the street is never in danger of forgetting that 
individuality is the supreme characteristic of human nature. It seems 
to him self-evident. But with the scientist the case is different. Of the 
several sciences devoted to the study of life-processes, none, pecul- 
iarly enough, recognizes as its central fact that life processes actually 
occur only in unified, complex, individual forms. Sciences find the 
very existence of the individual somewhat of an embarrassment and 
are disturbed by his intrusion into their domains. They pretend to 
deal with Nature, but are oblivious to the fact that Nature, as Goethe 
said, seems to have planned everything with a view to individuality. 


“Scientia non est Individuorut/i” 

Why is it that science and common sense part company over the 
fact of human individuality? The answer is that science is an arbitrary 
creed. It defines itself as a systematic attempt to trace order in nature 
through the discovery of regularities and uniformities characteristic 



of a whole class of objects. By choice, therefore, scientists have pre- 
occupied themselves with generalized truth, with occurrences that 
are common to events of one class. A “class,” to be sure, is a question- 
beggin g concep t, for it in turn is a n abstraction ~HeiIghed~to~c over 
co mmon occurrencesTSTlFYurns out that the “orde r in natu re’ ’ 
which the scientist seeks is after all quite a circu lar matter. 

The~or3er that is manifested in the single organism through the 
inter-relation of its bodily and mental processes is overlooked; it is 
not considered to be of legitimate scientific concern. The individual 
is regarded only as an instance or example of a universal principle; 
the search is always for broader and more inclusive formulations. “A 
description of one individual without reference to others may be a 
piece of literature, a biography or novel. But science? No.” 1 Scientia 
non est individuorum. 

There is a typical procedure the scientist feels compelled by con- 
vention to follow. He starts always with a certain professional atti- 
tude toward nature. The fact that this attitude is only one of many 
kinds of attitude of which he is capable, demonstrates at the outset 
a certain arbitrariness in his method of study. First, he makes a criti- 
cal discrimination of his subject matter, isolating from the individual 
who confronts him a chosen segment of behavior. This procedure is 
termed abstraction. He then observes the recurrence of this segment 
and its conditions in many members of a hypothetical class. Finding 
uniformity in the event and its attendant conditions, he makes a 
generalization or a law, and then, if he is a thorough investigator, he 
will submit his law to repeated tests and so establish it securely by 
empirical verification. 2 

The discovery of a law by this procedure is like finding a single 
thread running from individual nature to individual nature, visible 
only through the magical spectacles of a special, theoretic attitude. 
In everyday life, the scientist, like anyone else, deals effectively with 
his fellow men only by recognizing that their peculiar natures are 
not adequately represented in his discovery. The single functions 
which they have in common are deeply overshadowed by the indi- 
vidual use to which they put these functions. The piling of law upon 
law does not in the slightest degree account for the pattern of indi- 

1 M. Meyer, Psychol. Bull 1926, 23, p. 

2 These stages of scientific labor are 
scientific method; see, for example, A. W 


described repeatedly in treatises on the 
olf. Essentials of the Scientific Method. 



viduality which each human being enfolds. The person who is a | 
unique and never-repeated phenomenon evades the traditional scien- | 
tific approach at every step. In fact, the more science advances, the 
less do its discoveries resemble the individual life with its patent con- 
tinuities, mobility, and reciprocal penetration of functions. 

Starting with an infinitely more complex subject-matter than the 
other biological sciences, but with the same presuppositions, the psy- 
chologist has isolated his fragmentary elements, has generalized and 
verified his findings in the manner of the austere elder sciences. He 
has succeeded in discovering orderly processes in the “generalized 
mind,” but the phenomenon of individuality, so deliberately excluded, 
returns to haunt him. Whether he delimits his science as the study 
of the mind, the soul, of behavior, purpose, consciousness, or human 
nature,— the persistent, indestructible fact of organization in terms 
of individuality is always present. To abstract a generalized human . 
mind from a population of active, prepossessing, well-knit persons is 
a feat of questionable value. The generalized human mind is entirely w 
mythical; it lacks the most essential characteristics of mind,— locus,* ' 
organic quality, reciprocal action of parts, and self-consciousness. 

This exclusion of the individual from pure psychology has led to 
many anomalies. It has, for example, often been pointed out that the 
psychologist, in spite of his profession, is not a superior judge of 
people. He should be, but his ascetic and meager formulae derived 
from “generalized mind” do not go far in accounting for the peculiar 
richness and uniqueness of minds that are organic and single. The 
study of psychological laws is not sufficient training for the com- 
prehension of personal forms of mental life. Science is commonly 
considered to give men control over nature, but in the psychological 
field there is no “generalized mind” to be controlled. There are only 
single, concrete minds, each one of which presents problems peculiar 
to itself. In ordinary life we deal with our acquaintances, not by 
applying abstract laws, but by studying their individual natures. 

Still, with considerable tenacity, psychologists have held to con- 
vention, abstracting from minds initially organized such properties 
as suit their convenience, and their convenience is determined largely 
by scientific tradition. They are absorbed by the shadow of Method 
rather than by the individual objects upon which the shadow lies. 
To take a single example, the method of paired-comparison recom- 
mends itself as an objective and quantitative technique for studying 
judgments of the affective value of colors. In order to employ it the 



subject Is brought into the laboratory, and all variables other than 
color are controlled. He is shown two pure colored lights at a time, 
and states his preference. From a complete series of such judgments, 
the relative affective rank for each color is determined. Another sub- 
ject is called in, and then another. The goal is to find out how in the 
“generalized mind” one color takes precedence over others in affec- 
tive value. This particular attempt long ago had to be given up, for 
it was soon found that the generalized mind had no uniform prefer- 
ence. Individuals differed too markedly. Even the single individual 
has no constant affective response toward, say, green. Green light has 
a different value in the dark room of the laboratory and in a traffic 
signal; it has still different values as a property of an apple or a 
running brook, in a friend’s cravat or in an enemy’s automobile. Its 
pleasantness depends upon the observer’s interest, his memories, his 
mood. There is no affective value in abstract greenness. There are 
only personal experiences, and these determine the meaning and value 
of this contingent quality green. 

The founder of experimental psychology, Wilhelm Wundt, ad- 
mitted that “there is no psychological law to which the exceptions 
are not more numerous than the agreements.” 3 The exceptions, he 
recognized, result from the persistent intrusion of living individuality 
into the experimenter’s fragile abstractions. He then decided, with 
unassailable logic, that the direct study of this intruding fact is a 
necessary extension of psychological science. 

Now it is not to be doubted that here in the study of concrete 
individuals original talent and happy instinct must in the last analysis 
produce the best results, and that without these gifts a psychological 
analysis ... is impossible. But this does not mean that scientific con- 
sideration and practice would not be in a position to render essential 
service. . . . Such a combination of methods is the task of a practical 
psychology, namely, a characterology, which should investigate the 
basic and typical forms of individual character with the aid of prin- 
ciples derived from a general theoretical psychology, and a study of 
the relation and interaction of mental elements. Such a characterology 
Bacon has already demanded as a propaedeutic to politics and history. 
Unfortunately one cannot say that since Bacon any essential progress 
has been made in the task. But one may indeed prophesy that a solu- 
tion will depend above all else upon a full development of psychologi- 

B Phil. Studien y 1886, 3, 204* 


cal analysis and an overcoming of the one-sided intellectualistic and 

metaphysical tendencies in psychology . 4 

It is evident that the founder of experimental psychology per- 
ceived the dilemma, even though he did not offer a clear-cut solu- 
tion. Previously he had misappropriated the term “individual psychol- 
ogy” for what is today called general psychology. From individual 
psychology, therefore, curiously enough, he ruled out the study of 
the single individual, and consigned this important problem to “char- 
acterology.” 5 Although Wundt was right in his belief that general 
psychology has much to contribute to the study of single individuals, 
and right too in his statement that up to his time little progress had 
been made, he was nevertheless quite wrong in suggesting that the 
individual has no place in psychology proper, but only in a special 
science of characterology or “practical psychology.” He was wrong 
in holding that psychology should establish the laws, and character- 
ology account for the exceptions. 


More important than Wundt’s passing comments are those move- 
ments arising within the broad province of psychology as a protest 
against the prevalent neglect of the individual. Each of these move- 
ments, in greater or less degree, has attempted to improve the situa- 
tion, and has exerted marked influence: differential psychology, 
psychography , psychoanalysis, typology, Gestalt psychology, the 
psychology of Verstehen, purposive psychology, and personalistic 
psychology. Each deserves a critical note. 

Differential Psychology. In the early part of the nineteenth cen- 
tury astronomers created wide-spread interest by their accidental 
discovery of individual di fferen ces in reaction time. In the recording" 
of the time of the transit of stars, it appeared that there were innate 
differences in the speed with which the visual impression led to a i 
simple motor reaction such as the pressing of a key. Psychologists 
became concerned with the problem, and the study of the “personal 
equation” in reaction time entered Wundt’s laboratory . 8 Individual 
differences in reaction time, however, turned out to be only one kindj 

4 Logik, II, 4, i, (3), (b). The reference to Bacon is in the Advancement of 
Learning , VII, 3. 

5 See also, W. Wundt, Logik , II, 4, 2, (2), (a). 

6 E. G. Boring, History of Experimental Psychology, 1929, pp. 133-150. 



of individual difference. Everywhere the personal equation was 
found. Having commenced with reaction times, Wundt, with some 
reluctance, was compelled to admit the study of other individual 
differences. His reluctance, of course, came from his preconception 
of psychology as a science dealing with universal and not with indi- 
vidual characteristics of mental life. 7 Then came Galton with a totally 
different outlook. His primary interest lay precisely in the differ- 
ences between people, in their intelligence, imagery, and character. 
It is really Galton who deserves to be called the founder of differen- 
tial psychology. 8 

Differential psychology has become perhaps the most active 
branch of the science. Its attachments to the traditional outlook, how- 
ever, remain very close. Its method is by no means radical. Its first 
step is completely orthodox. It does just what general psychology 
does: selects a single attribute or function that can be conveniently 
isolated for study. It is concerned with the single attribute and not 
with the complex individual. The second step establishes the range 
and distribution of this attribute within the population of subjects 
employed. A third step is often added to discover the degree of co- 
variation between two or more of the functions or attributes thus 

In three important ways differential psychology fails to be an 
adequate method for the study of individuality. ( i ) Its interest, like 
that of general psychology, centers in the function or attribute that is 
isolated for study and not in the men possessing these functions. The 
individual is only a means, not an end. (2) The approach is as dis- 
tinctly elementaristic as in traditional psychology; it is “from be- 
neath” in terms of the elements of mind, and not “from above” in 
terms of their organization and patterning. In this respect differential 
psychology differs markedly from characterology. 10 (3) The impli- 
cation is that the sum-total of an individual’s scores on the isolated 
functions constitutes his individuality. The psychograph, with its 
separate plottings, is the utmost that the psychology of individual 

7 E. G. Boring, History of Experimental Psychology , 1929, p. 319. 

8 Ibid., p. 478. 

9 The chief tool of differential psychology is statistics. The range of individual 
differences in respect to a single variable is expressed usually in terms of mean or 
standard deviations. The co-variation of two simple variables is expressed by a co- 
efficient of correlation. When more than two variables are employed the principle 
of correlation is extended with the aid of such complex devices as multiple correla- 
tion, partial correlation, factor analysis, and the like. 

10 W. Stern, Differ entielle Psychologic, 3d edit., 1921, p. 12. 


differences has achieved in depicting the organization of mi n r! 
(cf. pp. 10-12). 

Mental tests are a typical achievement of differential psychology. 
Individuals are discovered to vary “normally” in some function (such 
as intelligence, perseveration, or introversion), and the degree of a 
person’s variation above or below the mean is considered to be his 
score. It is quite clear that the chief interest here is in one elementary 
attribute at a time. The peculiar patterning of attributes within the 
single person is not considered. 

Although the outlook of differential psychology is so limited, 
many psychologists are still inclined to regard it as the method par 
excellence for the study of individuality. Boring writes, “The psy- 
chology of individual differences because it deals with the particular 
and not with the general, tends to be practical.” But differential psy- 
chology does not deal with the particular at all, but rather with 
variations in the general. Dodge acclaims the great advance of differ- 
ential psychology over the older psychology of the “average” man: 
“Treating each individual as a special combination of capacities, ac- 
complishments and tendencies, has been far more productive than 
treating individuals as though they were all alike or as though they 
belonged to mutually exclusive types.” But actually, differential 
psychology has not at all treated the individual as a special combina- 
tion of capacities, accomplishments and tendencies. It has done 
nothing more than to imply that a person is a simple sum-total of his 
departures from the average . 11 

The reasons for the favored position of the psychology of indi- 
vidual differences are fairly obvious. In the first place it readily lends 
itself to standard experimental procedure, demanding only that 
enough cases be obtained to secure a reliable measure of the distribu- 
tion of differences. In the main it asks just what general psychology 
asks, that one function be abstracted at a time from the total complex 
person, and that these functions be observed and measured with 
accuracy. In the second place, it is a purely quantitative method, and 
rarely deals with more than one or two variables at a time. Although 
there is a tendency in recent years to develop mathematical tech- 
niques whereby more variables can be simultaneously treated, the 

11 The quotation from Boring is taken from his History of Experimental 
Psychology, p. 520; the quotation from Dodge, Conditions and Consequences of 
Human Variability, 1931, p. 6. This confusion concerning the province of differen- 
tial psychology is discussed by Allport and Vernon, Studies in Expressive Move - 
ment, 1933, p. viii. 



utmost that has been achieved still falls far short of the complete 
pattern of individuality. Even the most modem extension of differen- 
tial psychology, “factor analysis,” seeks to discover only that which 
is common. Individual distinctiveness arising from the arrangement 
and organization of these factors remains completely untouched. 

The method of group comparison is a branch of differential psy- 
chology. In the study of the psychological differences between the 
sexes, between races, or between two age-levels, it is customary for 
the investigator to compare the mean scores and the distribution of 
scores of the two groups in respect to one isolated function at a time. 
Here again the method lies, both in respect to its objective and its 
technique, well within the limits of traditional psychology. 

Figure i (modified from Stern) represents these methods. It is 
easy to see that both general psychology and differential psychology 
deal with abstracted attributes, considered common to all men, and 
represented on the horizontal dimension. The vertical dimension, 
where individuality is represented, is not served by the methods of 
either general or differential psychology. 

General psychology pursues the horizontal dimension, ideally 
without regard to individual differences, narrowing its object of in- 
vestigation wherever possible until variability is no longer a factor, 
or until it is found to take the form of a “chance” distribution. 

Differential psychology likewise deals with the horizontal dimen- 
sion, but is concerned with the range and dispersion of variation. One 
branch of this method is correlational psychology, another factor 
analysis, which characteristically follows through two, or sometimes 
more, functions (e.g., a, b, c, d). As a result the coexistence of these 
functions in the mental life of men may be demonstrated. By this 
method it may be show-n that certain functions are more commonly 
related in a population than certain other functions. But the patterning 
of the individual functions in the individual case is never directly 

The method of group comparison, a branch of differential psy- 
chology, is concerned with the differences between groups of subjects 
in respect to any one or more of the functions. The separation of the 
sexes in Figure i illustrates the possibilities in the field of sex differ- 
ences. Racial psychology employs the same method. 

Psychography. This movement is a direct offspring of differential 
psychology. It assumes that the individual is the sum-total of his 
scores on all the separate, measurable, psychological functions. In 

1 1 


Figure i, for example, Paul is represented by plotting his standings 
in respect to functions a, b, c, d, m, n, o, etc V This method is the 
last resort of those who believe in differential psychology, but who 
find that a single score in terms of deviation from the average f ails 

Examples of 

s ! 


PU . 1 









general intelligence 
mechanical aptitude 

rote memory 


auditory threshold 
red-green acuity 

muscle tonus 

affective range 
political prejudice 

a. . . . 




m. . . 

n. . . . 

0 . . . . 

u. . . . 

v. , . . 

w. . . 

X. . . . 


Figure i 

A Scheme for the Comparison of Methods 

to represent the individual. Having found nothing admirable in one 
unit, they try many units. 

Toulouse’s study of Poincare may be taken as an example . 13 
Although himself an experimental psychologist, Toulouse was not 
content with the study of mind-in-general, or as he said, of what is 
“common to an imbecile and to an Aristotle.” For this reason he 
undertook, in terms of deviations from the average, an analysis of the 
psychological functions, of a single individual. He chose the gifted 

12 Cf. Stem, op. cit., pp. 327-371. 

13 E. Toulouse, Henri Poincare , 1910. 



mathematician Poincare as his subject. He determined that Poincare 
had a memory span of eleven digits, that his associations with num- 
bers were prolific, that he had superior auditory imagery, and pos- 
sessed a relatively stable synaesthesia ( audition color ee), that he 
suffered from insomnia, had a fondness for music and not for hunt- 
ing, and that he seemed obsessed by his work. Surveying his efforts 
Toulouse admits that the genius of Poincare was somehow provok- 
ingly absent from his “synthesis.” 14 

Toulouse neglected especially the dynamic factors involved in 
the mental life of his subject. Interests, drives, values, and motives, 
complexes, sentiments, desires, and ambitions were slighted. A modern 
psychograph would be stronger for its inclusion of such material. 
But no psychograph escapes from the implication that the individual 
is merely the sum of separate functions which are studied in terms 
of deviations from the average. 

-Bs^kpanalysis. Unlike psychographic methods, psychoanalysis 
does not hold a quantitative view of mental organization. It seeks to 
understand patterns of desire and conflict. Since it deals in prolonged 
sessions with the examination of single persons it would seem at first 
sight to be a true science of individuality. Its method is clearly 
superior for disclosing the interlocking of dispositions and capacities 
in each individual life, and for tracing the dependence of present out- 
looks upon past experiences. 

But in spite of its valuable contributions (to be considered in 
Chapter VI) psychoanalysis fails to fulfill all the requirements of a 
science of individuality. ( i ) Like general psychology it is preoccu- 
pied with the search for universal causes. The properties of the un- 
conscious, it holds, are archaic and therefore the same for all people. 
The desires of the infant, his fixations, joys, fears, and the stages 
of development through which he passes are prescribed; the three- 
fold division of the self: the super-ego, the ego, and the id permit 
no variation; the behavior of people follows a conceptualized stand- 
ard and has essentially uniform significance. (2) Psychoanalysis is 
doctrinaire. The System is sacrosanct. Its design is traced upon the 
patient, and then —mirabile die tu— is discovered to exist there. The 
danger of fascination with theory is not unrecognized by psycho- 

14 Even before Toulouse undertook his psychography B. Perez attempted similar 
“synthetic” portraits, stressing qualities of temperament and dealing with ordinary 
Individuals rather than genius. His results were no more satisfactory. Cf. Le 
car act ere de P enfant a Phomme , 1892. 


analysts themselves. No doubt in their practice many of them indi- 
vidualize the lives of their patients more than they do in their theory. 
It is likewise common for one school of psychoanalysis to accuse 
another of slavery to creed; but it is not apparent which school is 
free to cast the first stone. (3) Psychoanalysis is not an eclectic move- 
ment; it has made virtually no contact with any other branch of 
psychology. The somewhat fantastic metaphors it employs show how 
little it has profited from the antecedent labors of psychological 
science. Much is known to general psychology about the unconscious 
operations of the mind, the processes of remembering, forgetting, 
dreaming, inhibiting, learning, reasoning, self-knowledge, and even 
about the mechanisms of motive and desire, and all of this knowledge 
psychoanalysis should, but does not, employ. No school of psy- 
chology can afford such splendid isolation. 

(4) The detachment of psychoanalysis from general psychology 
is due to its one-sided interest in the problems of psychopathology. 
Its doctrines have considerable pertinence in the study of the psycho- 
neuroses. Without modification, however, they are applied repeatedly 
to healthy mental processes; balance is interpreted in the same way 
as lack of balance; the sane are represented by the insane. The view 
that normality may be studied through the lens of abnormality is re- 
markably common, but it is none the less debatable. Even if it be 
sound, the equation should be reversible, and the study of normality 
should illuminate the field of abnormality. But this possibility is per- 
sistently overlooked. 

(5) Finally, the bulk of personal motives and traits which com- 
prise the individual are not, as psychoanalysis claims, necessarily 
rooted in the unconscious. They cannot all be understood simply by 
the art of deep-sea diving. Even where links are correctly traced 
between present trends and the experiences of childhood, they have 
often been so long rusted and broken that they are not, as analysts 
maintain, the bonds in the present structure of an individual’s life; 
in neurotics, perhaps, but in most people, no. Traits and interests, 
like plants, are capable of casting aside the shell of the seed from 
which they grew. Their direction of growth is upward into the 
future, and not downward into the past. In short, conscious motives 
and manifest behavior are of as great significance as are repressed 
motives and latent dispositions. 

Typologies. Any doctrine of types is a halfway approach to the 
problem of individuality, and nothing more. The typologist is ob- 



viously dissatisfied with a psychology of mind-in-general. He wishes 
to account for variety in human nature. It seems, however, that the 
task fatigues him, for he stops on the way somewhere between the 
abandoned “average” mind and the undiscovered fact that each indi- 
vidual is unique. It is difficult to disprove any given doctrine of type, 
particularly if “mixed types” are admitted, as they usually are. A 
type means nothing more than that certain people resemble other 
people in some respect. One may say, there are three types of people: 
those who use a convex tooth brush, those who use a concave, and 
those who use a straight tooth brush, plus, of course, the mixed type 
which uses sometimes one kind and sometimes another (or none at 
all). This is a valid typology if one is interested in tooth brushes. 
Similarly, there are people who are extroverted, those who are intro- 
verted, and those who are both. This, too, is true, if one is interested 
in extroversion and introversion. But the weakness lies in the narrow- 
ness of the interest. Should the investigator concern himself both 
with the habits of brushing teeth and with extroversion, he would find 
that a given individual belongs to more than one type. And if he is 
interested in all aspects of behavior, the individual turns out to have 
innumerable, unco-ordinated memberships. The patterning of these 
memberships within the person himself is overlooked. 

Figure 2 illustrates the fact that typological psychology has a 
tangential, and not a direct, bearing upon individuality. John, let it 
be supposed, is validly classified in four types. He resembles Messrs. 
A, B, C, D in physical features, and he resembles other individ- 
uals (E, F, G, H) in respect to his introversion, still others in 
imagery, and others in vocational interests. Each classification is 
correct, but John as an individual is almost untouched. His mental 
functions have been related to similar functions in other individuals, 
and have not been related within the organic field of his own nature 
to one another. Similarly every other individual escapes total inclu- 
sion; each is of a type only with respect to some one segment of his 
nature. Like general psychology, typology deals with abstracted 
attributes; its only advantage is that the abstracted attributes it selects 
are not regarded as universally distributed; only certain people have 

Here and there a typologist will argue that his schema is inclusive, 
that when John is fitted into his system, John’s position will be deter- 
mined not only for one mental function but for all. The typology of 
Kretschmer (cf. pp. 73-76), it is sometimes claimed, is complete. If 



John has such and such a physique, he must have such and such a 
temperament, together with a certain outlook upon life, a character- 
istic form of fantasy, and a particular manner of adjusting to the 
world. But such claims are made also for other typologies which do 

Typology vs. Individuality 
(Dotted ellipses signify types; solid circles, individuals) 

not overlap with Kretschmer’s at all. Typologies that are independent 
of one another cannot each be all-embracing. Every typological doc- 
trine tends toward extravagance. In reality types are valid only for 
a limited characteristic; they embrace a segment of individuality, but 
never the total individual . 15 

Gestalt Psychology. During the past twenty years there have been 
man y rebels within the ranks of general psychology attacking its 

15 Some of the most influential historical doctrines of type are described is 
Chapter III, and the critique of the concept is resumed in Chapter XL 



long-entrenched assumption that elements abstracted from individual 
experience are the proper data of the science. Not every opponent of 
elementarism, to be sure, is a Gestalt psychologist, but they all agree 
well enough that the time has come for a new emphasis upon the 
totalities and patterns of mental life. The special movement known 
as Gestalt psychology has four clear implications for the study of 

1. It deliberately shifts its emphasis from the traditional study of 
single segments of behavior abstracted from their natural setting to 
the study of the net-work of functions as they interpenetrate in the 
life of the organism. It may be instructive, writes Kohler, to study 
one hundred hearts together, but functionally considered a single 
heart has more in common with a pair of lungs than it has with other 
hearts. 16 Both general psychology and the psychology of individual 
differences have, as it were, studied hearts removed from their natural 
setting. A psychology of individuality will study the functions of an 
organism as they are related to other functions of the same organism, 
and not as they are related, through abstraction, to the activities of 
entirely different organisms. 

2. Another crucial innovation lies in Lewin’s distinction between 
genotypes and phenotypes. 17 The former are causal categories, the 
latter are descriptive. When similarity of appearance is observed in 
the behavior of different individuals the similarity in question may be 
said to constitute one phenotype, but not necessarily one and the 
same genotype. What appears to be the same effect may be due to 
radically different causes. For example, two individuals may have 
introverted natures (and thus belong to the same phenotype), but for 
very different reasons, one because of hereditary determination, and 
the other because of certain harsh experiences. Conversely, the same 
genotype, say an attack of infantile paralysis, may in the pattern of 
one individual’s life result in shyness, self-consciousness, and an atti- 
tude of defeat, in another the result may be compensation, vigor, and 
accomplishment. This distinction provides a more flexible treatment 
of individuality, for, according to it, the “same” cause, in the context 
of different lives, may produce contrasting effects instead of uni- 

3. Such an emphasis upon the individuality of motivational sys- 

18 W. Kohler, Gestalt Psychology , 1929, p. 351. 

17 K. Lewin, Gesetz und Experiment in der Psychologic . Reprinted from Sym~ 
posion, 1927, 5, 375-421. 



terns *s an improvement over the standard classification of impulses 
and drives advanced by psychoanalysis and by instinct-psychology. 
It provides also a corrective to the overemphasis upon the origin of 
traits. According to Gestalt psychology, the form of behavior may 
be determined by goals whose attainment lies in the future as well 
as by events of the past. Early habits are at best only one factor in a 
large system of forces determining at any given moment the form 
that individuality will take. 18 

4. A final bearing of Gestalt theory upon the study of individuals 
lies in its doctrine of the nature of understanding. Not only is the 
person well-structured, but he is perceived as well-structured. Ele- 
mentarism fails to account both for the organization of the individual 
mind and for the way in which it is understood by others. The 
argument has far-reaching significance, but it is too complex for 
discussion here. It will be considered in detail in Chapter XIX. 

The Psychology of Verstehen. This related school of thought is 
even more radical than Gestalt psychology. According to it there 
must be two separate and distinct psychologies, one to deal in the 
historic way with elements and the other to deal only with struc- 
tures; there is no common ground between them. The psychology 
of elements “destroys the meaningful totalities of life,” whereas the 
psychology of structures seeks deliberately the meaningful relation 
between the acts and experiences of the person. 19 The second method 
is the better, for it attempts to understand each unique mental life, 
without destroying its integrity. 

This is a drastic remedy for psychology’s ailment. Much would 
be lost by discarding altogether the methods of general and differen- 
tial psychology. Properly modified and adapted they can be employed 
in the study of individuality, and their past accomplishments in this 
direction are not altogether negligible. The problems of individuality 
are so complex that no legitimate method can be excluded. It would 
be foolish to repudiate the assistance of the psychology of elements 
wherever it may be accepted with profit. Psychology needs a 
broadening of outlook, not a schism. 

18 This virtue of Gestalt psychology turns out sometimes to be a limitation, for 
so complete is its emphasis upon “forces actually at work at a given moment” that 
it has difficulty in treating the consistency of the individual over a period of time. 
Disposition, latency, continuity, and the life-history are neglected problems. Even 
K. Lewin’s Dynamic Theory of Personality, 1935, deals almost entirely with the 
behavior of the individual in momentary (contemporaneous) “fields of force.” 

19 E. Spranger, Types of Men, trans. 1928, chap. L 



Purposive Psychologies. Wherever the conception of goal or pur- 
pose is paramount in psychology, there lies the implication that a 
personal life is involved. It is not, however, correct to assume that 
all dynamic psychologies are interested in the problem of individual- 
ity. In fact most of the doctrines of instinct, desire, need, propensity, 
attitude, or drive have been strictly within the tradition of the psy- 
chology of mind-in-general. Nevertheless, the purposive outlook 
makes the transition to individuality easier. Commencing in his 
Dynamic Psychology with an acceptance of the doctrine of common 
instincts, R. S. Woodworth arrives ultimately at a doctrine of the 
individuality of motive. Each person in the course of his life modi- 
fies his common hereditary equipment of impulses in ways that are 
peculiar to himself; he likewise develops new and autonomous inter- 
ests. With the acceptance of this doctrine the ground is obviously 
prepared for a dynamic psychology of the individual. This is a point 
of view of crucial significance and is developed at length in Chapter 
VII of this volume. 

Personalistic Psychology. The logical culmination of interest in 
the individual is the creation of a personalistic psychology. The chief 
tenet of this school of thought is that every mental function is em- 
bedded in a personal life. In no concrete sense is there such a thing 
as intelligence, space perception, color discrimination or choice re- 
action; there are only people who are capable of performing such 
activities and of having such experiences. It is improper to speak of 
the growth of skill, of vocabulary, or of knowledge; there is no 
growth excepting in the person; it is as part of his development that 
skill is enhanced, that vocabulary and knowledge are extended. Nor 
can motives ever be studied apart from their personal setting, they 
represent always the striving of a total organism toward its objective. 

It is an interesting fact that the most prominent representative of 
personalistic psychology, William Stern, was in past years the lead- 
ing figure in the study of differential psychology. It was in fact his 
perception of the limitations of the psychology of individual differ- 
ences, based upon long experience, that led him to attempt a more 
adequate formulation of the psychology of individuality . 20 Personal- 
istic psychology is still a relatively new school. Its guiding lines have 
been laid, but it awaits the rounding out that factual research alone 

20 W. Stern, Differentielle Psychologies 3d ed., 1921, p. 12; History of Psychology 
in Autobiography (ed., C. C. Murchison), Vol. I, 1930, pp. 360 f.; General Psychology 
from the Personalistic Standpoint , trans. 19^7. 



can give. Since, in a sense, this volume itself is a contribution to the 
same line of thought, further consideration may be postponed until 
all the evidence is assembled. The final chapter will return to the 


This chapter has dealt with a paradox. The scientific method, it 
is said, cannot study individuals. Yet within the province of scien- 
tific psychology several radical movements in the past fifty years have 
been focusing their interest with varying degrees of success upon 
the individual. Must it therefore be said that psychology, because of 
its growing interest in this subject, is becoming less and less scientific? 

Definitions are always arbitrary, even those in science. Therefore 
anyone who wishes to restrict the meaning of the sacred phrase, “the 
scientific method,” to the three-fold process of analysis, abstraction, 
and generalization is at liberty to do so. If the psychologist wishes 
to accept the same narrow definition, he too is free. But in doing so 
he must not transform what is merely an artifice of method into a 
doctrine of reality. Memory, intelligence, reaction, perception, sensa- 
tion, volition, and all like processes are real only in so far as they are 
organic; they are nothing more than attributes of personal activity. 
General psychology may, if it chooses, treat them as data, but the 
abstraction thus committed should be acknowledged. Titchener, a 
most ascetic methodologist, defined psychology as the study of ex- 
perience considered as dependent upon the experiencing person . 21 It 
would be impossible to devise a definition more satisfactory from the 
standpoint of personalistic psychology! But Titchener was so faithful 
to the narrower traditions of science that he did not appreciate the 
implications of his own definition. He sought only general laws of 
experience abstracted from the person. The logic of nature forced 
him to recognize the pre-eminence of the individual, but the conven- 
tions of science forced him to forget it. 

John Stuart Mill argued strongly that all definition in science must 
be “progressive and provisional,” for “any extension of knowledge 
or alteration in the current opinion respecting the subject-matter, 
may lead to a change more or less extensive in the particulars in- 
cluded in the science.” 22 At the present time current opinion in 

21 E. B. Titchener, Text-book of Psychology , p. 8. 

22 J. S. Mill, System of Logic } Book I, chap, viii, Sect. 4, 



psychology favors the inclusion of the phenomenon of individuality 
Therefore, no frozen definition should be allowed to interfere with 
the progress of investigation. Novel and somewhat daring methods 
will be required, and they should not be prejudged. The horizon is 
expanding. Yet there are no grounds for scientific dismay. The psy- 
chologists now bent upon the study of individuals are not likely to 
forget the rules they have learned in the laboratory for diminishing 
the risk of error in their observations. Simply because they wish a 
more liberal conception of science and of its methods, they do not 
therefore demand abolition of controls and a return to mysticism. 

Consider for a moment the method of experiment. It is the most 
serviceable technique of the exact sciences. At first glance it might 
seem inapplicable to the study of individuality. For one thing in 
order to secure suitable controls it is customary to isolate a simple 
segment of behavior for experiment, whereas individuality can never 
be secured a second time in the same way. The organism changes, 
and each direction of change influences all the interdependent func- 
tions. Doubt concerning the suitability of experiment for the study 
of the person has been expressed by Stern, 

The more exact [for general psychology] an experiment is— that is, 
the more elementary and isolated the phenomenon, and the more con- 
stant the conditions— the greater is its artificiality and the greater its 
distance from the study of the individual. Methods must above all else 
be suited to the problem in hand, and the problems set by general 
psychology and by the psychology of the individual are quite dif- 
ferent . 23 

An experiment, as everyone knows, involves exact observation 
and such control as admits of the isolation and the variation of con- 
ditioning factors. As a result of these requirements experiments fre- 
quently seem lebensfern, far removed from the study of the indi- 
vidual. But they need, not be. In recent years there have been satis- 
factory experiments dealing with complex levels of behavior— studies 
of gait and handwriting, vocal expression, personal style of work, 
and individual modes of thought, intuitive (non-analytical) judg- 
ments, and many other unorthodox subjects. In these studies, as 
Chapter XVI will show, the standards of experimentation have been 
well preserved, and yet the investigations conducted at a new level 
of complexity. If the experiments are not susceptible of exact repeti- 

23 W. Stern, Differ entielle Psychologic, p. 34. 



tion, neither, strictly speaking, are those of elementaristic psychology. 
Organisms continually change, and this fact must necessarily alter the 
behavior of elements as well as of patterns. 

To be sure, some problems of individuality completely elude the 
experimental method. One cannot approach experimentally such ex- 
periences as embarrassment, remorse, falling in love, or religious 
ecstasy. Wherever experiment is feasible it is to be preferred, where 
it is not, the problems still remain, and other methods must be em- 
ployed. Psychology is not exclusively experimental in its method, 
but neither for that matter is the eminently respectable science of 

Will a science treating individuals be able to evolve laws? The 
answer depends upon the conception of law. As law is ordinarily 
understood, it will not. If a law is a statement of an invariable associa- 
tion common to an entire class of objects, then, as Stern says, “indi- 
viduality is the asymptote of the science that seeks laws.” 

General laws have value in depicting the common ground upon 
which all individual minds meet. But this common ground is really 
a no-man’s land. When the investigator turns his eyes upon the indi- 
vidual he finds that in him all laws are modified, or as Wundt would 
have it, exceptions always occur. But a more liberal interpretation 
of the nature of law, considering it to be any uniformity that is ob- 
served in the natural order, is equally possible. In this sense, each 
person by himself is actually a special law of nature, so too is any 
structural occurrence within the pattern of his life. Though indi- 
viduality is never twice repeated, it represents nevertheless order in 
nature. If it were possible to grasp the complex totalities within a 
single individual life, to understand their formation, reciprocal action, 
directional tendencies, and dynamics— even though the discovery 
should have no wider application— it would be an achievement quite 
as significant as the establishment of any common law. 

As long ago as 1858, Samuel Bailey foresaw that the preoccupa- 
tion of psychology with mind-in-general would result in the neglect 
of the single individual. Accordingly he proposed that two sciences 
be recognized. 

In reference to the division concerning Individual and Personal 
Character, I may remark that it would be advantageous on several 
accounts to keep it distinct from Psychology, which, when confined 
to its proper objects, is chiefly occupied with describing, classifying 
and bringing under general laws, the phenomena of consciousness 



common to all mankind, and deals with Individual Character only 

incidentally and briefly— too briefly for the importance of the subject . 24 

Bailey’s advice that characterology be kept distinct from psychology 
has never been observed in English speaking countries. On the con- 
tinent, however, something of the kind has taken place. Since Bahn- 
sen’s work of 1867 a tradition of independent characterology has 
flourished. But its independence of psychology has unfortunately 
meant a greater dependence upon guess-work, figurative discourse, 
and esoteric metaphysics. The advice to study “personal character” 
without the aid of a general and experimental psychology is dark 

The proposal to distinguish sharply between the study of general 
principles and the study of the individual case has taken many other 
forms. The philosopher Windelband, for example, proposed to sepa- 
rate the nomothetic from the idiographic disciplines. 25 The former, 
he held, seek only general laws and employ only those procedures 
admitted by the exact sciences. Psychology in the main has been 
striving to make of itself a completely nomothetic discipline. The 
idiographic sciences, such as history, biography, and literature, on the 
other hand, endeavor to understand some particular event in nature 
or in society. A psychology of individuality would be essentially 

The dichotomy, however, is too sharp: it requires a psychology 
divided against itself. As in the case of the two psychologies (the 
analytical and the descriptive) advocated by Dilthey and Spranger, 
the division is too drastic. It is more helpful to regard the two methods 
as overlapping and as contributing to one another. In the field of 
medicine, diagnosis and therapy are idiographic procedures, but both 
rest intimately upon a knowledge of the common factors in disease, 
determined by the nomothetic sciences of bacteriology and bio- 
chemistry. Likewise, biography is clearly idiographic, and yet in the 
best biographies one finds an artful blend of generalization with indi- 
vidual portraiture. A complete study of the individual will embrace 
both approaches. Half a century ago this same conclusion was 
reached by the French psychiatrist, Azam. The science of charac- 
ter, he wrote, “cannot proceed by generalities, as does psychology, 
nor by individualities, as does art. It occupies an intermediate posi- 

24 S. Bailey, Letters on the Philosophy of the Human Mind, i8jj-j8, II, 265. 

25 w. Windelband, Geschichte und Naturwissenschaft, 3d edit., 1904. See also, 
R. Eisler, W drier buch der philosophischen Begriffe, 1904, p. 512. 


2 3 

cion.” 20 One should now add that this “intermediate position” will 
fall properly within the scope of a broadened psychology. 

The psychological study of individuality will continue in the 
manner of general psychology to employ experimentation, not on the 
simple level to be sure, but on the level of traits, interests, and the 
personal idiom. It will be interested in the laws of learning and in all 
genetic principles, but will attempt to co-ordinate these in the nexus 
of individuality. It will employ critical standards of observation, 
avoid the impressive single instance, and profit from all other hard- 
earned lessons of psychological science. But the interest will be 
broader. It will embrace the problem of intra-individual consistency 
as well as inter-individual uniformities; it will not be content with 
the discovery of laws pertaining to mind-in-general, but will seek also 
to understand the lawful tendencies of minds-in-particular. But there 
is no need for two disciplines. Psychology can treat both types of 
subjects. Its position will be stronger for having enlarged its hori- 
zon. The individual has now taken his place within this horizon, and 
he has come to stay. To study him older methods will be adapted 
and new ones invented, as exact as the subject will permit. 

If there are psychologists who in the face of this growing move- 
ment still declare that the study of the individual is not and never 
can be a part of science, they must now be left alone with their 
views. The psychological study of individuality is being undertaken 
with profound seriousness; no blind loyalty to an anachronistic ideal 
can prevent it. One may call it science or not science, as one chooses. 
Long before the method of natural science attained its commanding 
position with psychology paddling in its wake, there was an ancient 
meaning of Scientict. It prescribed no method; it set no limits; it signi- 
fied simply knowledge. 

26 E. Azam, Le caractere dans la smite et dans la maladie , 1887, p. vL 



But this word persona has rolled along with wonderful 
bounds, striking right and left, suggesting new thoughts, stir- 
ring up clouds of controversy, and occupying to the present 
day a prominent place in all discussions on theology and phi- 
losophy, though few only of those who use it know how it 
came to be there. 

— F. Max Muller 

In the previous chapter the term individuality was used to signify 
the separateness and uniqueness of each human being. But mere 
separateness and uniqueness are not the psychologist’s chief concern. 
Wasps and mice, trees and stones possess this elementary distinction. 
In addition to separateness and uniqueness a human being displays 
psychological individuality, an amazingly complex organization com- 
prising his distinctive habits of thought and expression, his attitudes, 
traits and interests, and his own peculiar philosophy of life. It is the 
total manifold psycho-physical individuality, commonly referred to as 
personality, that engages the attention of the psychologist . 1 

The term “personality” expresses admirably this interest of the 
psychologist, and yet it is a perilous one for him to use unless he is 
aware of its many meanings. Since it is remarkably elastic, its use in 
any context seldom is challenged. Books and periodicals carry it in 
their titles for no apparent reason other than its cadence, its general 
attractiveness, and everlasting interest. Both writer and reader lose 
their way in its ineffectual vagueness, and matters are made much 
worse by the depreciation of the word in the hands of journalists, 

1 The question whether animals have personality in the psychologist’s sense 
should probably be answered in the affirmative, for animals like people manifest 
individuality in their patterns of habit and expression. See, F. Schwangart, Jahrbuch 
der Charakterologie, 1928, 5, 101-140; also T. Schjelderup-Ebbe, in A Handbook of 
Social Psychology ? 1935, chap, xx; D. Katz, “Characterology and Animal Psychol- 
ogy,” Univ. Maine Stud 1930, No. 14, pp. 28-59. Nevertheless, the mental indi- 
viduality of lower animals is too rudimentary to serve as a prototype of human 
personality, and for that reason will not be considered in this volume. 



beauty doctors and peddlers of gold bricks labeled “self-improve- 

“Personality” is one of the most abstract words in our language, 
and like any abstract word suffering from excessive use , 2 its connota- 
tive significance is very broad, its denotative significance negligible. 
Scarcely any word is more versatile. 

Let us take such a word as Person. Nothing can be more abstract. 
It is neither male nor female, neither young nor old. As a noun it is 
hardly more than what to be is as a verb. In French it may even come 
to mean nobody. For if we ask our concierge at Paris whether any- 
body has called on us during our absence, he will reply, “Personne, 
monsieur,” which means, “Not a soul, sir.” 3 

The only way to re-vitalize such a frayed concept is to trace its 
history. There is no single correct definition of “personality”; usage 
has sanctioned too many. Some of the meanings are psychological, 
some are not. The first task is to distinguish between them. The sec- 
ond is to select from among available psychological definitions one 
that best fits the phenomenon dealt with in this volume. 

There is probably no single word of greater interest to philolo- 
gists. To bring within a few pages the fruits of their investigations is 
a hazardous undertaking, but the only practicable alternative is to 
omit entirely historical conceptions, thereby isolating the present day 
psychology of personality from the history of human thought, a very 
common, but nevertheless very undesirable, practice. 

The terms personality in English, persomialite in French, and Per- 
sonlichkeit in German closely resemble the personalitas of medieval 
Latin. In classical Latin persona alone was used, but its meanings are 
in so many respects equivalent to those of the modern terms that our 
historical sketch must commence with persona itself. 


Persona originally denoted the theatrical mask first used in Greek 
drama and adopted about a hundred years before Christ by Roman 
players. (Legend has it that a popular Roman actor was responsible 
for the importation to hide his unfortunate squint.) The Greek desig- 
nation for mask was prosopon, a word having a vague resemblance to 

2 Thorndike lists personality in the eighth thousand of English terms arranged 
in order of frequency. E. L. Thorndike, The Teachers’ Word Book , 1921. 

*F. Max Muller, Biographies of Words, 1888, pp. 32 f. 



persona. Several authorities regard the Latin term as a direct deriva- 
tion from the Greek . 4 Critics of this theory point to the improbability 
of such a marked alteration occurring in the form of the word. To 
escape this difficulty other philologists favor its derivation from peri 
soma (around the body), and still others from the Etruscan and Old 
Latin word persum (head or face) . 5 Some trace its origin from the 
Latin per se urn (self-containing) . e 

The antecedent of persona favored by most authorities, however, 
is the Latin phrase per sonar e (to sound through) J According to this 
theory the term had reference to the large mouth of the mask or 
perhaps to a reed device inserted into it for projecting the voice of 
the actor. Those who hold this view concede that from the very 
beginning persona, by a metonymic change, referred not so much to 
the vocal aspects of the mask as to its visual properties. 

Whatever its antecedents may have been, no philologist denies 
that persona at one time designated the theatrical mask. But drama 
and life, the actor and the role, the assumed and the real character, are 
too intimately related to withstand confusion. In rapid succession all 
within classical times, a series of extensions and transformations took 
place, converting this concrete noun into one that is abstract and 
multiple in meaning. In the writings of Cicero (106-43 B *c.), prob- 
ably not long after the word first appeared, are found at least four dis- 
tinct meanings of persona: 8 

(a) as one appears to others (but not as one really is) ; 

(b) the part someone ( e.g ., a philosopher) plays in life; 

(c) an assemblage of personal qualities that fit a man for his work; 

(d) distinction and dignity (as in a style of writing) . 

4 See H. Rheinfelder, “Das Wort ‘Persona,’” Zsch. f. roman. Fhilol. , 1928, 
Beiheft 77, pp. 22 if. 

5 These various theories may be readily traced through Rheinfelder, op. cit. 3 
and Muller, op. cit . 

6 While this derivation is generally rejected it is of interest to observe its 
emphasis upon the unitary nature of the person. Cf. S. Schlossmann, Fersona und 
Upoa&TTov im Recht und im christlichen Dogma , 1906, p. 13 ftn. 

7 Among the writers accepting this derivation are: A. Trendelenburg, “Zur 
Geschichte des Wortes Person,” Kantstudien, 1908, 13, pp. 4 iff.; Muller, op. cit.; 
J. R. Greenougli and G. L. Kittredge, Words and Their Ways in English Speech , 
1902, p. 268. 

There is only one significant objection to this theory, namely, a conflict in 
vowel quality. In per sonar e the o is short, whereas in persona it becomes long. 
Such alteration in the quality of root vowels is somewhat rare, although, as Muller 
points out, by no means unknown. 

8 Cf. Miiller, op. cit., pp. 38 f* 



The first meaning, of course, has the original significance of the 
mask; the second suggests real status, not the mere pretence; the third 
signifies the inner psychic qualities of the player himself; and the final 
meaning connotes importance and prestige which later were con- 
veyed by the derivative term personage. 

These early transformations are summarized in Figure 3. Starting 
with Persona as Mask, the four earliest derivative meanings are listed 

Figure 3 

The Early Derivative Meaning of Persona 

and numbered. Each serves as a radiant point for later semantic 
change. From them it is possible to derive all of the fifty types of 
definition listed in this chapter. Nearly all of these definitions are still 
in current use as accepted meanings for person or personality. It 
would be impossible to claim that the lines of connection here drawn 
are in every detail historically accurate. What is intended is a chart of 
relations of meaning rather than an exact chronology of change. 

The first logical extension of Persona as Mask refers then to 

external appearance (not the true self). (1) 

The adjective personatus meant provided with, or wearing, a mask. 
Unlike persona it never took on the meaning of the true self. It was 



always fictitious, counterfeit, a pretense. The meaning is preserved 
today in Jung’s doctrine of “persona” and in some popular definitions 
of personality (see p. 42). 

The second root meaning signifies the 

character or role which the player assumes in the drama. (2) 

In German one may hear the phrase “Er hat seme Person gut 
gespielt .” This meaning is likewise preserved in the still current phrase 
dramatis personae. The theater-goer of today is, however, not quite 
certain whether the personae are listed in the right or in the left 
column of his program; he does not know whether the parts in the 
drama are indicated, or whether the personal names of the actors are 
intended, or both. This confusion was possible even in ancient Rome, 
for persona often indicated the player himself, considered as 

an individual possessed of distinctive personal qualities. (3) 

This meaning, of greatest importance in the development of psycho- 
logical definitions, appears as the third usage in Cicero’s writings. 

The fourth derivation, likewise found in Cicero, has the signifi- 
cance of 

prestige and dignity. (4) 

This meaning was eagerly assimilated into the Roman caste system, 
wherein some individuals had legal rights and obligations, and others 
did not. Hence persona was used to indicate 

the free bom citizen (as distinguished from the slave). (5) 

Still in classical Latin, the term came to mean 

a representative (one who stands for a group or insti- , 
tution), (6) 

whence the Latin expression persona grata. The representative of the 
church came later to be called its 

parson (Anglo-Saxon, persoun). (7) 

Important persons, possessed of prestige and dignity (4) came also 
to be known as 

personages. (8) 

As a latter day example, the family name McPherson derives from the 
cluster of meanings expressed in definitions (5), (6), (7), and (8). 



2 9 

A by-path was created by the Latin grammarian Varro (nd- 
2 6 b.c.) who early spoke of the 

persons in grammar, (9) 

a metaphor based upon the resemblance between dramatic and syn- 
tactical roles. 

These definitions, as well as all those that follow, can be ordered 
according to their position in a continuum of meanings ranging from 
the external (false, mask-like) manner to the internal (true) self. This 
double and contradictory reference is the outstanding characteristic 
of the term persona, and of the contemporary term personality as 
well. Personality means that which is assumed, non-essential, false, as 
well as that which is vital, inward, and essential. Which meaning the 
psychologist should choose will be discussed later. 


The choice of certain Church Fathers who used the term to 

the Members of the Trinity 


markedly advanced the equivalence between persona and the inner 
(true) self. Many writers have speculated upon the remarkable trans- 
formation of a word which originally meant assumed manner, and 
came to have the opposite significance of inner nature, substance, and 
even essence. 9 

By the third century a.d. the elements of the Trinity were desig- 
nated as Personae. At first there seemed to be danger here of an 
heretical error; for did not persona signify variable appearance, a 
mask-like pose? Would not a Deity assuming three alternative roles 
be forced to abandon two of them while representing the third? That 
such heresy was avoided is due no dout>t to the flexible connotation 
of the term itself which made it possible to think of substance while 
speaking of masks; the Deity indeed subsisted in three Persons, but 
each shared the same essence. Although the matter was thu? settled to 
the satisfaction of the Church Fathers, later theologians continued for 

9 Cf. C. Webb, God and Personality, 1919, e.g., p. 20 and p. 36; A. Trendelen- 
burg, op, cit ,, p. 12; R. Hirzel, “Die Person/’ Sitz. d, kon , Bayer . Akad . d, Wissen - 
schaften , 1914, 10, p. 53. P. Caros, Personality , 19 u, p. 18; H. Rheinfelder, op, ctU 
p. 165. 


long to discuss the paradox of varied appearance and single sub- 
stance . 10 


Through associating the concept of Person with true essence the 
early theologians helped prepare the way for the celebrated definition 
given by Boethius in the sixth century: 

Persona est substantia individua rationalis naturae. ( 1 1 ) 

Taking the substantial nature of the person for granted Boethius adds 
the attribute of rationality, thus breaking ground for a long line of 
subsequent philosophical definitions of personality. 

The definition given by Boethius was satisfactory to virtually all 
the philosophers of the Middle Ages. St. Thomas Aquinas exalted the 
person above every other reality observed in nature; nothing, he 
thought, was superior in dignity to the beings who possess rational 
individuality. 11 The cultivation of personality was not explicitly em- 
phasized in medieval ethics, but inasmuch as man should live a life that 
is rational, the growth of personality was clearly an implied good. 12 
In this manner the emphasis was gradually taken from Aristotle’s 
belief that the individual existed for the good of the species, and came 
to be placed upon respect for the integrity and value of the indi- 
vidual. This trend reached its culmination in later romanticism and in 
personalistic ethics. 

Christian Wolff emphasized as the chief criteria of the person 

self-consciousness and memory. 12 (12) 

This view is not unlike that of Leibnitz who defined a person as a 

substance gifted with understanding. 14 (13) 

10 Cf. S. Schlossmann, op. cit. 

11 For the place of personality in Scholastic philosophy of value see E. Gilson, 
“L’esprit de la philosophie medievale,” chap, x of Le personnalmne chretien , 1932. 

12 Classical Latin employed only persona , personalis (adj.) and personaliter 
(adv.) . The substantive personalities is not found until medieval Latin (Rheinf elder, 
op. cit., pp. 155 ff.). It was a characteristic of nouns based upon the adjectival root 
and formed in this way, that they gained somewhat in abstractness. The difference 
in meaning, however, between personalitas and persona should not be exaggerated. 
The break in the form of the word does not correspond to an equally sharp change 
of meaning, 

is “persona dicitur ens, quod memoriam sui eonservat, hoc est, meminit, se 
esse Idem iilud ens quod ante in hoc vel isto fuit stato” Psychologia rationalis, 1734, 
sec. 741, p. 660. 

14 G. W. Leibnitz, Hauptschriften zur Grundlegung der Philosophie (ed. by 
E. Cassirer), 1906, Vol. II, p. 184. 


Locke stressed still more the attribute of self-consciousness. 15 A per- 
son for Locke is 

a thinking, intelligent being, that has reason and reflec- 
tion and can consider self as Itself. (14) 

In more recent times Windelband has reinforced Locke’s definition. 
Personality is 

individuality which has become objective to itself. (15) 

This capacity is possessed in varying degrees by different individuals, 
so that one man may possess more personality than another. 18 

Rickert returns to the purely metaphysical conception of person- 
ality as essence. 17 The true personality is to a large degree 

the indivisible center, and only in the periphery do the 
processes of alteration take place. (16) 

Since these philosophical definitions with their varying emphasis 
upon rationality, self-consciousness, and the subjective core of a man’s 
being, are indistinguishable from various parallel conceptions of the 
Self, some writers explicitly equate personality with 

Selfhood. 18 (17) 

Personeity was used by Coleridge in the nineteenth century as equiva- 
lent to selfhood, but the term did not take root. 

Other philosophical definitions consider personality as an ethical 
rather than a metaphysical conception. There are writers, for ex- 
ample, who regard it as the ideal and perfect attribute of Being, never 
fully attained by human kind. To Lotze, for example, personality was 

the ideal of perfection (18) 

reached only by God though approached in varying degrees by 
men. 10 This is likewise Webb’s view: personality is “suggested by 
what we find in men” but “only imperfectly realized in them.” 20 

15 An Essay Concerning Human Understanding , Book II, chap, xxvii, sec. 9. 

16 w. Windelband, An Introduction to Philosophy, trans. 1921, p. 281. 

17 H. Rickert, Die Grenzen der naturmssenschaftlichen Be griff sbildung, 3d ed* 
1921, p. 237. 

18 Cf. Roberta Crutcher, Personality and Reason, 1931, p. 75. 

19 H. Lotze, Microcosmos, 4th ed., 1897, p. 682. 

20 Webb, op. cit., p. 21. 

3 2 


Also one psychologist, William Stem, has maintained that the fully 
fashioned personality is that at which men aim but never fully attain. 21 
The significance of this doctrine lies in its emphasis upon the con- 
tinuous growth and development of all personal characteristics. The 
personality is always striving for greater completeness in accordance 
with its own ideals. This point of view might be phrased as 

the person being formed by unattained ideals ( 19) 

The conception of personality as an ideal is exalted still further in 
Romanticism. 22 Goethe, for example, regarded personality as the 

supreme value. 23 (20) 

Like Goethe, Nietzsche and William von Humbolt, who spoke often 
of personality, would not have man sacrifice his integral, entire self 
to any one part of his nature. 

A somewhat more sedate expression of the ethics of personal in- 
tegrity is found in Kant. “Everything in creation, except one thing, 
is subject to the power of man, and can be used by man as a means to 
an end; but man himself, man the rational creature, is an end in him- 
self. He is the subject of moral law and is sacred by virtue of the 
autonomy of his individual freedom. . . . 

Personality exhibits palpably before our bodily eyes the 
sublimity of our nature.” 24 (21) 

The so-called ethics of self-realization, placing higher value upon 
the development of the individual personality than upon any other 
goal, has, of course, a Kantian origin. One illustrative definition will 
suffice. “Personality is 

21 William Stern, Die menschliche Personlichkeit , 1923, p. 20. 

22 For an exposition see T. J. McCormack, “Personality, A Study in the History 
of Verbal Meanings,” Me 72 t. Hygiene , 1931, 15, 34-44. 

23 “Volk und Knecht und Uberwinder, 

Sie gestehn zu jeder Zeit: 

Hochstes Gltick der Erdenkinder 
Sei nur die Personlichkeit.” 

W est-ostlicher Divan , Buch Suieika. 

24 j Kritik der praktischen Vernunft , Kant's gesammelte Schriften , Bd. V. (Reim«i 
Verlag), 1908, p. 87. 


that quality in every man which makes him worth 
while, aside from the uses to which he may be put by 
his fellows.” 25 ( 22 ) 

Kantian ethics is commonly considered likewise to be a starting 
point for personalistic doctrines. Some of these take a voluntaristic 
turn. Thus Hetherington and Muirhead define personality as 

“that form of individuality . . . which is rendered pos- 
sible by the possession of mind and will. To be a person 
is to be one and indivisible, but it is a unity that is 
achieved, not by the suppression of natural instincts, 
temperament, and capacities, but by the permeation of 
them with a common spirit— the power of finding free- 
dom, not frovi them but in them.” 26 (2 3 ) 

Personalistic philosophies agree (a) that the personality is of su- 
preme value, (b) that persons are to be distinguished metaphysically 
from things, and (c) that subjective experience is the final psycho- 
logical court of appeal. According to Bowne, “personality can never 
be construed as a product or compound; it can only be experienced 
as a fact.” 27 

“The essential meaning of personality is selfhood, self- 
consciousness, self-control, and the power to know. 

These elements have no corporeal significance.” 28 (24) 

On the other hand, Stern, though a personalist, does not deny cor- 
poreal significance to personality, but declares the person to be “psy- 
chophysically neutral,” 

a multiform dynamic unity. 29 (25) 

The personalistic position has importance for psychology in so 
far as it focuses attention upon the organization of the individual 
mental life; but its significance becomes metapsychological when, for 
example, it treats institutions, nations, or God as “persons.” This 
super-individual realism is illustrated in Kant’s doctrine that 

25 M. F. Adler, in Essays in Honor of John Dewey, 1929, p. 8. 

The ethics of self-realization is expounded by James Seth, A Study of Ethical 
Principles, rev. ed., 1904; J. Dewey and J. H. Tufts, Ethics, 1908; R. C. Cabot 
The Meaning of Right and Wrong, 1933. 

26 H. I. W. Hetherington and J. H. Muirhead, Social Purpose, 1918, p. 104. 

27 B. P. Bowne, Personalism, 1908, p. 264. 

28 Ibid., p. 266. 

29 W. Stern, Die menschliche Personlichkeit, p. 4. 



“A person is a rational being with rights; if he has 
duties too, he is a man; if not, he is a God.” 30 (26) 


Figure 4 

Some Inter-relations of the Philosophical Meanings of Personality 

These varied philosophical conceptions of personality may 
greatly enrich the psychologist’s view of his problem, even though 
in many instances they extend far beyond his professional sphere. 
In particular, definitions (11), (12), (14), (19) and (25) suggest the 
30 C. C. J. Webb, Kant's Philosophy of Religion , 1926, pp. 181 f. 


attributes of rationality, self-consciousness, conative striving, and 
absolute uniqueness that ought surely find their way into the psy- 
chologist’s store. With the metaphysical and ethical conceptions he 
will have less to do. Figure 4 summarizes the expansion of meaning of 
personality in the hands of theologians and philosophers. Again it 
must be said that the connecting lines in the diagram do not neces- 
sarily represent historical currents of influence but serve primarily to 
call attention to similarities or extensions of meaning. 


It was established by Justinian that a slave was not a person. Slaves 
were held to be a-personal. Only freeborn men had the dignity of 
persons; see Definition (5). It was against such social discrimination 
that the Christian moralists raised their cry. Each man, they insisted, 
is a person. Although yielding finally to this moral suasion, the juristic 
conception of the nature of the person began with the Roman Code. 
Since, according to this Code, only freeborn citizens could be bearers 
of rights, claiming the privileges and protection of law, it followed 
that a person was 

any individual enjoying legal status. (27) 


In time the' individual’s material possessions with which the law like- 
wise had concern came to be known as his personalty. 

In modern times, in civilized countries, all individuals may claim 
the protection of law, and hence the standard legal conception of the 
nature of the person has broadened out, showing clearly the influence 
of Christian moral doctrine: today a person at law is “any being 
having life, intelligence, will, and separate individual existence, dis- 
tinguished from an irrational brute and inanimate thing; a human be- 
ing, as including body and mind; an individual of the human race; a 
living person, composed of body and soul; a man, woman, or child; 
a moral agent; a self-conscious being; the whole man.” 31 This gen- 
erous definition makes person equivalent to 

the living human being in his entirety. (28) 

In the course of its development the law came to embrace not 
only the rights and duties of individual human beings, but of whole 

si Corpus Juris (ed. by W. Mack & D. J. Kiser), 1929, Vol. 48, sec. i, pp. 1037 f. 



groups of people, of “corporations.” There grew up therefore the 
conception that a person might be 

an incorporated group of people. (29) 

Such a group, to be sure, is generally designated as an “artificial per- 
son,” 32 though there are some who argue that corporate groups are 
“real” enough from a philosophical point of view. 33 

Closely related to this juristic conception of the “group person” 
are many of the group-mind doctrines found in idealistic philos- 
ophy. An illustration may be taken from Royce: 

“A genuinely and loyally united community which 
lives a coherent life is in a perfectly literal sense a 
person.” 34 (30) 

Similar is the statement quoted by a newspaper from an address given 
by Nicholas Murray Butler, president of Columbia University, 
“Each nation is a moral personality with a mind and heart and soul.” 
Such definitions extend the concept of personality far beyond the 
individual-organic locus with which the psychologist prefers to deal. 
Nevertheless it is necessary to note these trans-individual definitions, 
for they serve as a background to the numerous sociological concep- 
tions next to be reviewed. The preceding juristic definitions as well 
as the sociological definitions that follow are summarized in Figure 5. 


Definition (28) held that person may mean simply an individual 
human being. When we ask, for example, “How many persons were 
there?” nothing more is intended than an inquiry concerning the 
number of living beings who were present. Person in this sense means 
then a human “unit of mass.” More especially since the word person- 
ality came into use (in English in the fourteenth century) the term 
person has been freed for this simple usage. Oddly enough, as Max 
Muller has pointed out (p. 25), the same term in French means not 
only a single human being, but likewise 

82 Beals defines a corporation as an “artificial person, created by law as an 
entity independent of the natural person or persons composing it, and endowed 
by the law that creates it with the power of acting as such independent person.” 
J. H. Beals, The Law of Foreign Corporations , 1904, p. 7. 

83 Cf. F. Hailis, Corporate Personality, 1930, p. 240. 

34 From a letter to Mary W. Calkins printed in Papers in Honor of Josiah 
Royce , 1916, p. 67. 


no human being. (31) 

Here is no doubt an instance of what philologists call “condensa- 
tion,” and comes about through the fact that in the expression ne 
personae the ne is phonetically lost. 

Many times emphasis is placed upon the physical portion of the 
self, as in such phrases as “to injure the person” or “to expose one’s 
person.” One definition, therefore, must be in terms of 

the bodily self. (32) 

A variant of the same definition places special emphasis upon 
physical presence, as in the phrase “will appear in person.” This ex- 
pression once seemed redundant, but it is less so now in the days of 
cinema and radio when partial appearance or appearance not in per- 
son is possible. 

A curious blending has appeared between Definition (28)— the 
single human individual— and an inversion of Definition (4), weight 
and dignity. The result is 

an expression of contempt (33) 

as in the phrase “That person! ” In this case it seems almost as though 
an individual were scarcely human; if a person at all, then certainly 
devoid of the dignity and worth that a true personality would possess. 

The term personality like “person” has also one connotation of 
disparagement, as in the expression “Do not indulge in personalities,” 
spoken when the listener wishes to interrupt slander or incrimination. 
This meaning is frequently encountered in eighteenth century Eng- 
lish literature. It might be defined as 

revealing qualities of others in offense to good taste. ( 34) 

Proceeding again from Definition (28) we find that sociologists, 
thinking always of society at large, regard the person, not merely as 
an individual being, but, to use Eubank’s expression, 35 as 

the ultimate granule of the human group. (35) 

It is characteristic of all sociological definitions of personality that 
they deny to it the attribute of self-sufficiency. In one way or an- 
other personality is always considered a reflection of, or dependent 

35 E. E. Eubank, “The Concept of the Person,” Sociol. & Soc. Res., 1927, 12, 
p. 363. 



upon, the social ground. One succinct and fairly typical statement 
says that personality is 

the subjective side of culture. (36) 


Figure 5 

Juristic and Sociological Meanings 

This view, expressed by Paris 36 and others, overlooks completely the 
part played by biological determinants (by intelligence, tempera- 
ment, and physical heredity) and considers the subjectification of the 
36 E. Faris, “The Concept of Social Attitudes,” J. Appl. Sociol. , 1925, 9, 404-409. 


customs and social traditions within a single human life as the whole 
story. It is obviously a one-sided view. 

There are other, better balanced, sociological definitions; some in 
fact are intricately synthetic: the following definition by E. W. 
Burgess, for example, 37 

Personality is the integration of all the traits which 
determine the role and status of the person in society. 
Personality might, therefore, be defined as social effec- 
tiveness. (37) 

This definition seems to contain remnants of all the original Ciceronic 


Definitions of personality in terms of the outer appearance of a 
man, that is, in terms of his “stimulus value” for others, are as common 
as they are troublesome. They have, to be sure, the virtue of etymo- 
logical fidelity, for of all the various types of definitions they cling 
most closely to the original meaning of mask, signifying 

deceptive masquerade or mimicry. (38) 

Cicero, in one context of his writing, asked, “Why should I walk 
around like a persona?”— why, that is, should he assume an appearance 
false to his nature, pretending to be that which he was not? It has 
already been shown that persona means not only what a man is (3), 
but likewise precisely the opposite: what he is not (1)! This am- 
biguity has never been overcome even in modern definitions of per- 
sonality. In Latin it was somewhat lessened in certain contexts 
through the adoption of the derivative participle personates to mean 
exclusively wearing a false appearance. In English we have the verb 
“to personate”; more commonly, “to impersonate.” 

Now, in everyday life it is often necessary to mask the true self, 
to present to the world an appearance of conventional fitness. For 
this common practice of wearing a conventional mask Jung has pre- 
served in his system of psychology the original term persona, defin- 
ing it as 

37 Contained in Proc. of the Second Colloquium on Personality Investigation 
(Johns Hopkins University Press), 1930, p. 149. 

4 o 


“a mask of the collective mind, a mask that disguises 
individuality ... a stage part spoken by the collective 
mind.” as (39) 

Many of the popular definitions of personality are of the same 
type as Jung’s definition of persona , and it is here that confusion 
arises. Is personality a solidly organized system of traits and senti- 
ments or is it the mask of a poseur presented to a critical world? Is it 
the moi pro-fond et essentiel of Bergson, or is it the moi superficial? 39 

Colloquial speech, influenced by the idols of the theater and the 
market-place, equate personality with charm, with “It,” with 

superficial attractiveness. (40) 

A cosmetic advertisement claims that a certain lipstick will confer 
“personality” upon its user. In this instance personality is not even 
skin-deep! 40 

Popular definitions have two serious defects. First, they refer only 
to some portion of the intricate pattern of personal life, generally to 
the vitality, expansiveness, or expressiveness of the individual Sec- 
ondly, they invariably consider personality only in terms of its in- 
fluence on other people, and never in terms of its subjective or in- 
terior organization. 

This second deficiency is found frequently in the writings of psy- 
chologists who embrace the biosocial view of personality. The bio- 
social view stands in sharp contrast to the biophysical conception 
represented in this volume, which holds that personality, psychologi- 
cally considered, is what an individual is regardless of the manner in 
which other people perceive his qualities or evaluate them. 

As an example of the biosocial view, two definitions will serve. 
One considers personality to be those “habits or actions which suc- 
cessfully influence other people.” 41 According to this narrow defini- 
tion the secret schemes, frustrations, worries, and private aspirations 
that never become socially effective are not parts of personality. 
Another definition speaks of “the sum-total of the effect made by an 
individual upon society.” 42 These views would deny personality to 

38 C. G. Jung. Die Beziehungen zwischen dem Ich und dem Unbewussten , 1928, 
P- < 54 * 

39 H. Bergson, Essai sur les donnees hnmediates de la conscience , 19th ed., 1920, 
pp. 97-106. 

40 For a discussion of popular definitions see A. A. Roback, Personality r the 
Crux of Social Intercourse , 1931, chap. i. 

41 E. G. Flemming, /. Educ. SocioL, 1933, 7, p. 409. 

42 H. C Link, The Return to Religion , 1936, p. 89. 


4 1 

the solitary hermit or to Robinson Crusoe (at least before the advent 
of Friday). They would also invite a perilous distinction between 
“more” and “less” personality, for it is obvious that individuals have 
various degrees of social effectiveness. A queen of the movies seen by 
millions of people on the screen would have incomparably “more” 
personality than the complex and tortured poet dwelling in attic 
obscurity. In contrast, the biophysical view would argue that each 
human being has “as much” personality as any other. The effective- 
ness of a life is indeed a problem (best handled in social psychology), 
but it is by no means a criterion for the existence of personality. A 
solitary dweller on a desert island, unknown to any other mortal, has 
a full-fledged (and intensely interesting) personality. 

The biosocial views are best summarized in the brief definition 
proposed by May. According to him personality is a man’s 

social-stimulus value. (41) 

“It is the responses made by others to the individual as a stimulus 
that define his personality.” 48 

In favor of this view it is sometimes urged that only through 
the judgments of other people are our personalities known at all. As 
we appear to others so must we be, for there is indeed no other 
criterion of our natures. A man’s personality is what others think of 
him; it is, in short, his reputation. 

If this reasoning be sound then it might as truthfully be said that 
a fish, a tree, or a star can be defined only in terms of their “social 
stimulus value,” for is it not through their stimulus-giving properties 
that the scientist comes to know them? Philosophers have repeatedly 
made this same point in insisting that the scientist’s own perception is 
ultimately the definitive test of existence. But it is none the less cer- 
tain that the scientist safely infers that fish, trees and stars do exist 
outside of his own mind. Should he persist, because of his errors of 
sense or of judgment, in regarding the fish as an eel, or the tree as 
twice too large, or the star as a ball of cheese, the essential nature of 
these objects would still remain unchanged. It is so with personality. 
Our only basis for knowing people is their stimulus-properties. But 
our errors of judgment and perception do not change their personali- 
ties any more than a star becomes a savory because it is misperceived 

<8 M. A. May, “The Foundations of Personality,” chap, iv of Psychology at 
Work (ed. by P. S. Achilles), 1932. 



by the scientist who has a taste for cheese. If it is here objected that 
the criterion for truth is always social, that the greatest agreement 
among the greatest numbers is the only possible test, the proper reply 
is that whatever holds for stars holds also for personality. Personality 


Figure 6 

Biosocial Definitions (in Terms of External Appearance) 

is just as objective a fact as any other event in nature; there are no 
special grounds to single it out for biosocial definition. 

It is not helpful to the psychologist therefore, to assume that per- 
sonality is merely an impression made by one individual upon others. 
Unless the investigator approaches personality as directly as he would 
any other objective event he will be swamped by considerations of 
reputation, rumor, gossip, erroneous evaluations, and social effective- 
ness. If the biosocial view were followed to its logical conclusion it 
would not be necessary to study the perceptions, reactions, prejudices 
and interests of the personality vn question at all, but rather the per- 


ceptions, reactions, prejudices and interests of everyone else in his 
social circle ! 44 


Even in classical times persona had the meaning of an assemblage 
of personal qualities (3). In English by the seventeenth century this 
usage was firmly established. It is this tradition more than any other 
that serves as background for all current biophysical definitions of 
personality. It contains no reference to drama, to pretense, or to mere 
reputation. It is, nevertheless, a broad conception that allows a num- 
ber of varying points of stress. Indeed it is rare to find one psycholo- 
gist who will accept another’s definition, even though both agree that 
personality is a biophysical event. In spite of the multiplicity of defi- 
nitions it is possible to assort them into five basic classes. 45 

1. 0 )nnibus Definitions. Perhaps the commonest type of defini- 
tion is the one that starts with the phrase, “Personality is the sum- 
total of . . .” In place of “sum-total” we sometimes find equally 
nondescript expressions such as “composite,” “aggregate,” “ensem- 
ble,” “congeries,” or “constellation.” Typical examples are: “the sum- 
total of the reactions of an individual to all the situations which he 
encounters”; 46 or “a constellation of the following event patterns— 
somatic reactions, autistic reveries, adjustive thinking, and object 
orientations.” 47 One of the best known omnibus definitions is that 
of Prince: “Personality is 

the sum-total of all the biological innate dispositions, 
impulses, tendencies, appetites, and instincts of the indi- 
vidual, and the acquired dispositions and tendencies— 
acquired by experience.” 48 (42) 

44 Sensing the paradox in this position May adds to his biosocial view the state- 
ment that it “does not deny the reaction side but insists that the stimulus side be 
included in the picture” and adds that “fully 90%” of research investigations deal 
with personality as an objective fact, i.e., “take the reaction point of view” (op. cit., 
pp. 84 f .) . It appears then that the biophysical position taken in this book is ac- 
ceptable not only to the majority of psychologists, but even, in part, to Professor 
May himself. 

45 For a somewhat fuller treatment of these five classes see G. W. Allport and 
P. E. Vernon, “The Field of Personality,” Psychol Bull, 1930, 27, pp. 681-687. 

46 l. G. Lowrey, in Proc. Second Colloquium on Personality Investigation, 
1930, p. 151. 

47 H. D. Lasswell, ibid., p. 151. 

48 Morton Prince, The Unconscious , 2d ed., 1924, p. 532. 



The confusion resulting from this view of personality as an omnibus 
piled high with all manner of unarranged impedimenta can be seen 
in Menninger’s definition— which seems almost satirical. He writes, 
“Of course personality is used to describe almost everything from the 
attributes of the soul to those of a new talcum powder. As we shall 
use it, it means the individual as a whole, his height and weight and 
loves and hates and blood pressure and reflexes; his smiles and hopes 
and bowed legs and enlarged tonsils. It means all that anyone is and 
all that he is trying to become.” 49 

Such omnibus definitions render absolutely no service to science. 
They are glib and reckless, and at best define merely by enumeration. 
They omit the most outstanding phenomenon of all mental life, 
namely, the presence of orderly arrangement. The mere cataloguing 
of ingredients defines personality no better than the alphabet defines 
lyric poetry. 

2. Integrative and Configurational Definitions. Unlike the omni- 
bus definitions this second class stresses the organization of personal 
attributes. A simple formulation of this type is that of Warren and 
Carmichael: 60 

the entire organization of a human being at any stage 
of his development. (43) 

A more complex definition still placing primary stress upon organiza- 
tion, but a secondary stress upon distinctiveness or uniqueness, is that 
of MacCurdy: B1 

an integration of patterns (interests) which gives a 
peculiar individual trend to the behavior of the or- 
ganism. (44) 

Similar is the definition of A. Gesell: “the pervasive superpattern 
which expresses the integrity and the characteristic behavioral indi- 
viduality of the organism.” 02 

3. Hierarchical Definitions. These are known by the demarcation 
of various levels of integration or organization, usually with the image 
of a capstone, or innermost self, to dominate and to center the pyra- 

49 K. Menninger, The Human Mind , 1930, p. 21. 

60 H. C. Warren and L. Carmichael, Elements of Human Psychology , 1930, 
P- 333 * 

51 J. T. MacCurdy, Common Principles in Psychology and Physiology , 1928. 
p. 263. 

52 In Proc . Second Colloquium on Personality Investigation , 1930, p. 149. 



mid of personal life. The prototype for conceptions of this sort is 
to be found in James’s classic treatment of the four levels of the 
Self. 53 Self is essentially the personality “viewed from within.” 
(James’s preference for the Self was not unnatural in the introspective 
era in which he wrote. He used the term personality only in referring 
to the phenomena of dissociation, hysteria, and multiple personality 
popularized by the French school.) 54 

There is first the material Self, including the body, one’s posses- 
sions and one’s family and friends whom one cherishes. There is next 
the social Self determined by the recognition one obtains from one’s 
associates. Regarding this level James made his famous statement that 
a man “has as many different social selves as there are distinct groups 
of persons about whose opinion he cares.” 55 As though feeling that 
he had here dangerously dismembered the personality, James hastened 
to the third level, the spiritual Self that unifies so far as is possible 
man’s discordant tendencies. A fourth level, The Pure Ego (the 
knower, the Self of Selves), is required by certain philosophical sys- 
tems, but is not, according to James, psychologically distinguishable 
from the third level. With this hierarchical scheme as a model many 
authors have treated personality in a similar way, McDougall, for in- 
stance, and Bridges, Heider, Blondel, Martin, and many others. 08 All 
of them offer their conception in terms of 

levels or layers of dispositions, usually with a unifying 
or integrative principle at the “top.” (45) 

4. Definitions in T enns of Adjustment. Biologists and behaviorists 
are inclined to view personality as an evolutionary phenomenon, as a 
mode of survival. According to them personality is the “whole-organ- 
ism-in-action.” The point of view is most fully developed by Kempf 57 
whose conception, in essence, is 

the integration of those systems of habits that represent 
an individual’s characteristic adjustments to his environ- 
ment. (46) 

There is so much that is manifestly true and important in this con- 
ception that it will receive fuller discussion in Chapters IV and V. 

63 William James, Principles of Psychology, 1890, Vol. I, chap. x. 

M Cf. Th. Ribot, The Diseases of Personality, 2d ed., trans. 1895. 

55 Op, cit., p. 294. 

56 See G. W. Allport and P. E. Vernon, Psychol. Bull., 1930, 27, p. 684. 

67 E. J. Kempf, “The Autonomic Functions and the Personality,” Nerv. <b 
Ment. Dis. Monog., 1921, No. 28. 

4 6 


5. Definitions in Terms of Distinctiveness. Schoen writes, “If all 
the members of any one social group acted alike, thought alike, and 


Figure 7 

Psychological Meanings 

felt alike, personality would not exist,” and hence proposes as his 

“Personality is the organized system, the functioning 
whole or unity of habits, dispositions and sentiments 
that mark off any one member of a group as being dif- 
ferent from any other member of the same group.” 58 (47) 

58 M. Schoen, Human Nature , 1930, p. 397. 


A similar definition is Wheeler’s, “that particular pattern or balance 
of organized reactions which sets one individual off from another.” SB 

An early and somewhat singular use of the term to indicate dis- 
tinctiveness in the functioning of a single organ, is Mitchel’s phrase 
“absolute personality of the eye,” employed in connection with the 
long-standing problem of the “personal equation” or individual dif- 
ferences in reaction time. 60 

Another variant of definitions of this type is Woodworth’s. 61 
This author believes that every act of the individual is colored by 
personality. Personality is not substantive, it is adverbial; it is the style 
of life. “Personality refers not to any particular sort of activity, such 
as talking, remembering, thinking or loving, but an individual can 
reveal his personality 

in the way he does any of these things.” (48) 

These psychological definitions thus far considered are summar- 
ized in Figure 7 which serves likewise as the frame of reference for 
Definition (50) offered as the guide for the author and reader of 
this volume. 


From this long survey of past and present usage what may we 
conclude? Since there is no such thing as a wrong definition of any 
term, if it is supported by usage, it is evident that no one, neither 
the theologian, the philosopher, the jurist, the sociologist, the man 
in the street, nor the psychologist, can monopolize “personality.” For 
the psychologist, to be sure, some definitions seem to be more service- 
able than others. Completely unsuitable are biosocial formulations in 
terms of social reputation or superficial charm (40) and (41). The 
distinction between reputation (social effectiveness) and the true per- 
sonality is one that will be observed rigidly throughout this book. 
Omnibus definitions (42) must likewise be rejected. More helpful 
are those conceptions that ascribe to personality a solid organization 
of dispositions and sentiments. Valuable likewise are definitions that 
refer to the style of life, to modes of adaptation to one’s surroundings, 
to progressive growth and development and to distinctiveness, 

59 R. EL Wheeler, The Science of Psychology, 1929, p. 34. 

60 Cf. E. G. Boring, A History of Experimental Psychology , 1929, p. 14a 

61 R. S. Woodworth, Psychology, 1929, p. 553. 


Might we not merely say that, psychologically considered, per- 
sonality is 

what a man really is? (49) 

This terse expression states the essential biophysical position, and is 
acceptable enough in principle. Yet it is too brief and vague as it 
stands. The following amplification seems to serve the purpose better: 


This formulation contains the seeds of the hierarchical, integrative, 
adjustive, and distinctive classes of definitions described above. In a 
sense, therefore, it represents a synthesis of contemporary psycho- 
logical usage . But each portion of the definition is chosen for a par- 
ticular reason, and these reasons must be made clear if the definition 
is to be accurately understood. 

Dynamic Organization. To escape from the sterile enumerations 
of the omnibus definitions it is necessary to stress active organization. 
The crucial problem of psychology has always been mental organiza- 
tion (association). It is likewise the outstanding problem dealt with 
in this volume. Hence “organization” must appear in the definition. 
Yet this organization must be regarded as constantly evolving and 
changing, as motivational and as self-regulating; hence the qualifica- 
tion “dynamic.” Organization must also imply at times the correlative 
process of disorganization , especially in those personalities that we are 
wont to regard as “abnormal.” 

Psychophysical Systems. Habits, specific and general attitudes, sen- 
timents, and dispositions of other orders are all psychophysical sys- 
tems. In later chapters these dispositions will be ordered within a 
theory of traits. The term “system” refers to traits or groups of traits 
in a latent or active condition. The term “psychophysical” reminds us 
that personality is neither exclusively mental nor exclusively neural. 
The organization entails the operation of both body and mind, in- 
extricably fused into a personal unity. 

Determine . This term is a natural consequence of the biophysical 
view. Personality is something and does something. It is not synon- 
ymous with behavior or activity; least of all is it merely the impression 
that this activity makes on others. It is what lies behind specific acts 
and within the individual The systems that constitute personality are 




in eveiy sense determining tendencies , and when aroused by suitable 
stimuli provoke those adjustive and expressive acts by which the per- 
sonality comes to be known. 

Unique . Strictly speaking every adjustment of every person is 
unique, in time and place, and in quality. In a sense, therefore, this 

Figure 8 

Summary View of the Definitions of Personality 

criterion seems redundant. It becomes important, however, in our 
later discussions of the problem of quantitative variation among indi- 
viduals in respect to the so-called “common” traits (see Chapter XI), 
and is therefore emphasized in the definition. 

Adjustments to His Environment . This phrase has a functional and 
evolutionary significance. Personality is a mode of survival. “Adjust- 
ments,” however, must be interpreted broadly enough to include mai- 

5 ° 


adjustments, and “environment” to include the behavioral environment 
(meaningful to the individual) as well as the surrounding geographical 

Above all, adjustment must not be considered as merely reactive 
adaptation such as plants and animals are capable of. The adjustments 
of men contain a great amount of spontaneous, creative behavior to- 
ward the environment. Adjustment to the physical world as well as to 
the imagined or ideal world— both being factors in the “behavioral 
environment”— involves mastery as well as passive adaptation. 


Figure 8 relates by number all of the fifty representative defini- 
tions contained in this chapter. The connecting lines, it must once 
more be repeated, indicate only meaningful relations, and not neces- 
sarily historical sequences. The reader will observe that in general 
the more “outer,” more superficial definitions fall to the left, the more 
“inner,” metaphysical definitions fall to the right. The definition to 
which this volume will adhere (50) represents the convergence of 
many trends of semantic change. It is perhaps as nearly central and 
synthetic in its significance as any single definition could be. It repre- 
sents a distillation of much of the speculative thought in past ages 
and of much of the scientific research in recent times. 


Character is a term frequently used as a synonym of personality. 
It has a history as long and nearly as intricate. As employed by 
Theophrastus it possessed much of the same adverbial significance 
that Woodworth ascribed to personality. It was the “engraving” of 
the individual, his style of life as determined by his dominant trait 
(see pp. 56 ff.). There is no historical reason why the term should 
not be used interchangeably with personality, as indeed it frequently 
is. But in modem psychology there are two divergent lines of mean- 
ing both of which give an independent significance to the term, 
thereby lessening the practice of equating the terms. 62 

Many writers identify character with some special phase of per- 

02 For various meanings of character see L. Klages, The Science of Character . 
transl. 1929, pp. 38 ff.; W. McDougall, “Of the Words Character and Personality,” 
Char . & Pers., 1932, 1, pp/3-16; also J. B. Watson, Psychology from the Standpoint 
of a Behaviorist, 1919, p. 392 ftn. 



sonality, making it a subdivision of the whole. For example, it is said 
that personality may be viewed as intelligence plus character, 63 or as 
intelligence, temperament, and character. 64 Since personality is never 
an additive phenomenon such statements serve to characterize neither 
personality nor character. 

Whenever character is considered to be a subdivision of person- 
ality, it is nearly always identified with volition in some way; thus, 
“the degree of ethically effective organization of all the forces of the 
individual,” 65 or “an enduring psychophysical disposition to inhibit 
impulses in accordance with a regulative principle.” 66 Large numbers 
of writers hold this view; for all of them character is the aspect of 
personality that engenders stability and dependability, that is respon- 
sible for sustained effort in the face of obstacles, or works for remote 
ends rather than those that are nearer in time but of less worth. This 
is the meaning endorsed not only by many psychologists but by the 
church, by educators, and by common speech as well. 

With due respect to the prevalence of this usage one must still 
question the wisdom of separating the volitional faculty from the 
remainder of personality. Activities involving “will” emanate from 
the most complex systems of personality, but not from any one area 
that can be arbitrarily designated as character. When a man shows 
“character” by resisting temptation, or when it is said that the aim 
of education should be the “development of character,” what is really 
meant is that the man has behaved, or the child should be trained to 
behave, in ways that are approved by prevailing social and ethical 
standards. The exercise of “will” in each case is a phenomenon of 
personality. Character enters the situation only when this personal 
effort is judged from the standpoint of some code. If will-power is 
shown in a man’s behavior, then will-power is in his personality; if 
constancy, inhibition, self-respect, the power of “prolonging the 
vestibule of desire,” or of “keeping the selected motive dominant 
throughout life” characterizes his behavior, then these talents are 
important features of his personality. Social standards as well as psy- 
chology are brought in when we label such conduct “character.” 

On all sides one encounters confusion of psychology with ethics. 

The layman asks the psychologist, “How ought I bring up my child?” 

63 G. G. Femald, J. Abnorm. Psychol., 1920, 15, p. 1. 

64 E. Kahn, Psychopathic Personalities, 1931, p. 32. 

85 W. S. Taylor, J. Abnorm. & Soc. Psychol., 1926, 21, p. 86. 

68 A. A. Roback, The Psychology of Character, 1927, p. 450. 

5 2 


And the psychologist is presumptuous enough to tell him; although no 
psychologist qua psychologist can tell how a child ought to be brought 
up. The most he can do is to disclose human nature as it is, and then, 
after a moral code has been chosen , find out means of incentive and 
training that will achieve the end desired. Unreflectingly many psy- 
chologists, particularly mental hygienists, pose as experts in ethics, for 
do they not speak with assurance about the desirability of “mental 
adjustment”? Strange to say, it is often the Behaviorist, the most 
“rigidly scientific” in his ideals for psychology who traffics most un- 
guardedly in ethics. The child “should” grow up without complexes 
and conflicts; he “ought” not have fears. Such “guidance” is not psy- 
chology at all; it is pure ethics, springing from an uncritical accept- 
ance of the normative ideal of “perfect adjustment.” Were the psy- 
chologist not guided by this particular code (popularized by mental 
hygiene), but rather by an ethics of perfectionism, self-realization, or 
rationalism (to name only a few alternatives) he would have to recast 
completely his advice and prescriptions. 

Therefore, instead of defining character as the volitional aspect of 
personality, it is sounder to admit frankly that it is an ethical concept. 
Sir John Adams writes, “Character is the moral estimate of the indi- 
vidual, an evaluation.” Defined in this way, the psychologist does not 
need the term at all; personality alone will serve. Character is per- 
sonality evaluated, and personality is character devaluated . Since 
character is an unnecessary concept for psychology, the term will not 
appear again in this volume, excepting in quotations from other 
writers, or in a clear historical context . 67 

Biologists are still able to employ the term as meaning a distin- 
guishing attribute, for example, in the phrase “inheritance of an ac- 
quired character.” If it were not for the unhappy invasion of psy- 
chology by ethics this simple and pristine meaning would still be 
serviceable. One would speak, for example, of the “characters of a 
personality,” meaning, o.f course, distinctive determining tendencies. 
Although this usage is now unfortunately impossible, the word char- 
acteristic remains unspoiled, both as an adjective and as a substantive. 
It is an admirable word to express the graven marks of individuality 
and can be employed in psychology without ethical connotation. It is 
especially serviceable as a generic term to cover habits, traits, attitudes, 
and interests, in so far as they are distinctive determining tendencies 

67 In the following chapter, for instance, it is convenient to employ the term 
characterology in its broad historical sense, as the science of the characteristics of 



of the individual. It is a curious fact that characteristic should have 
kept its primitive and psychologically useful meaning, whereas its root- 
form has gathered so much ethical moss. 


The classical doctrine that ascribed peculiarities of temperament 
to the humors of the body has persisted throughout the ages so that 
the meaning of the term has varied but little. The term found its way 
into English in the Middle Ages along with the doctrine of the 
humors. It meant then and still means a “constitution or habit of 
mind, especially depending upon or connected with physical consti- 
tution.” Today in America psychological writers stress particularly 
the constitutional basis; for them temperament is the “internal 
weather” in which personality develops; it is the subjective climate 
provided by native physiological and kinetic endowment. The usage 
of the term in Great Britain is somewhat different, tending to equate 
temperament with personality, as in the phrase “temperament tests” 
(rather than “tests of personality”) . 

Temperament, like intelligence and physique, might be said to 
designate a certain class of raw material from which personality is 
fashioned. Strictly speaking there is no temperament apart from per- 
sonality, nor any personality devoid of temperament. It is merely 
convenient to employ the term in speaking of dispositions that are 
almost unchanged from infancy throughout life (dispositions satu- 
rated with a constant emotional quality, with a peculiar pattern of 
mood, alertness, intensity, or tonus) . The more anchored a disposition 
is in native constitutional soil the more likely it is to be spoken of as 

It is seldom doubted today, any more than it was among the an- 
cients, that temperament is dependent somehow upon the biochemical 
constitution. Work dealing primarily with glands, physical build, or 
blood composition (to name only a few popular fields of contem- 
porary study) frequently claims to be seeking the biological founda- 
tions of personality. And so it is— indirectly; but first of all it is seek- 
ing the physical correlates of temperament. There is danger of exag- 
geration in labels such as “Glands Regulating Personality,” “Bio- 
logical Foundations of Personality,” “Physique and Character.” The 
implication in these titles is that no factors other than the constitu- 
tional need be considered. It would clarify the problem if biologists 



and endocrinologists would drop the term personality altogether, and 
speak exclusively of temperament. It is in this latter field of research 
that their efforts and those of the psychologist can be most directly 

The following definition of temperament fits standard psycho- 
logical usage and meets the requirements of this book. T empenmient 
refers to the characteristic phenomena of an individual's emotional 
nature , including his susceptibility to emotional stimulation, his cus- 
tomary strength and speed of response, the quality of his prevailing 
mood, and all peculiarities of -fluctuation and intensity in mood; these 
phenomena being regarded as dependent upon constitutional make-up, 
and therefore largely hereditary in origin. 



The term Characterology as employed in this chapter refers to all 
the diverse schemes advanced in the past to account for, or to depict, 
forms of human individuality. The word itself was first employed by 
J. Bahnsen in 1867. 1 But in spite of its relative recency it serves very 
well to designate all the forerunners of the modem psychology of 

The fact that each human being is unique and distinctive has al- 
ways caused astonishment and comment, provoking some of the 
greatest thinkers throughout the ages to theorize and to observe. 
Employing as they did the language of pre-scientific psychology 
some of their formulations now seem quaint and old-fashioned. But 
our efforts today are faltering enough, so why should we dismiss 
these venerable contributions with amusement or condescension? 

The reader will find in A. A. Roback’s Psychology of Character 
(1927) a rich record of past descriptions of human types, and of 
methods, postulates, plans and projects for the discovery of the basic 
principles of the development and activity of personality. 2 While 
there is no reason to attempt another complete history of the subject, 
certain authors emerge from the past with whose contributions read- 
ers of this book must be familiar. It is convenient to classify these 
outstanding figures as representative of six different schools of 
thought. Three have their sources in ancient Greece, viz., literary 
characterology , hwnoral psychology, and physiognomy; three are of 
modern origin: phrenology, ethology, and experimental character- 
ology . Within these movements we shall pay particular attention to 

1 J. Bahnsen’s work is in two volumes, entitled, Beitrdge zur Charakterologie , 
and was republished in 1932. For a discussion of the term, see W. Stern, Differ en - 
tielle Psychologies 3d ed., p, 11, and E. Utitz, Charakterologie , 1925, p. 8. 

2 Likewise a chronological table of the leading events in the history of charac- 
terology will be found in A Bibliography of Character and Personality compiled 
by the same author in 1927. This volume also contains 3,300 classified references 
to literature in the field. A shorter history of characterology is offered by J. Jastrow, 
Pop . Set. AIo. } 1915, 86, 590-615. 



the contributions having lasting value, and pass lightly over their 
erroneous by-paths, 


“Character writing,” as it is technically called, is a minor literary 
form originating in Athens. Some say Aristotle invented the form, for 
proof pointing to his characterization of the Magnanimous Man in 
the Nichomachean Ethics. Others hold that the originator of the 
form was Theophrastus, Aristotle’s pupil and successor at the Lyceum. 
Whoever the actual inventor, it is certain that Theophrastus’ crisp, 
brilliant, detached representations of human types brought him last- 
ing fame and established a model for innumerable imitators in the past 
two thousand years. 

A successful “Character” is a brief descriptive note which so 
aptly depicts a common type of human being that it is recognized 
and appreciated by the readers of any age and in any land as a sim- 
plified but essentially correct image. Descriptions of particular indi- 
viduals are not Characters; they are Portraits. Actually, of course, 
Characters and Portraits shade into one another, and certain Portraits, 
notably those of La Bruyere, possess such universality of application 
that they belong with the productions of Theophrastus. Strictly 
speaking a Character is a type, drawn by the accentuation of some 
dominant disposition or trait. The 30 Characters of Theophrastus 
follow a rigid style of composition. Each commences with a brief 
definition of the dominant trait, and continues with typical instances 
of the operation of this trait. The numerous imitators of Theophrastus 
depart in varying degrees from this rigid formalism, so much so at 
times that they virtually forsake the tradition of Character writing. 

It is a peculiarity of Theophrastus’ Characters that all present some- 
what vicious or at least unpleasant types of personality. Whether he 
found the good man too dull to write of, whether he wished to hold 
up to the Athenians their own mean and sordid conduct, or whether 
the 30 sketches which survive are only a portion of those written, are 
all possibilities for conjecture. The text of his work suffered consider- 
ably from Byzantine interpolation but has now been purified, and 
several good translations are available. 2 It was part of the Byzantine 

8 R. C. Jebb, The Characters of Theophrastus, 1909; R. Aldington (ed. and 
trans.), A Book of Characters, E. P. Dutton and Co., 1925; C. E. Bennett and William 
A. Hammond, The Characters of Theophrastus, 1902; Francis Howell, The Charac- 
ters of Theophrastus, 1824. (The last two have physiognomical sketches.) 



tradition that Theophrastus wrote his Characters when he was 99 
years old. 4 It is not difficult to find in the shrewd observations and 
detachment of the sketches reasons for assigning them to a writer of 
venerable age and rich experience. The following example (from 
Aldington’s translation) demonstrates the style of Character writing 
in the hands of the Greek master. Though written 2,200 years ago it 
is applicable to some of our acquaintances today. 


Penuriousness is economy carried beyond all measure. A Penurious 
Man is one who goes to a debtor to ask for his half-obol interest be- 
fore the end of the month. At a dinner where expenses are shared, he 
counts the number of cups each person drinks, and he makes a smaller 
libation to Artemis than anyone. If someone has made a good bargain 
on his account and presents him with the bill he says it is too much. 

When his servant breaks a pot or a plate, he deducts the value from 
his food. If his wife drops a copper, he moves furniture, beds, chests 
and hunts in the curtains. If he has something to sell he puts such a 
price on it that the buyer has no profit. He forbids anyone to pick a 
fig in his garden, to walk on his land, to pick up an olive or a date. 
Every day he goes to see that the boundary marks of his property 
have not been moved. He will destrain on a debtor and exact com- 
pound interest. When he entertains the members of his deme, he is 
careful to serve very small pieces of meat to them. If he goes market- 
ing, he returns without having bought anything. He forbids his wife 
to lend anything— neither salt nor lamp- wick nor cinnamon nor 
marjoram nor meal nor garlands nor cakes for sacrifices. “All these 
trifles,” he says, “mount up in a year.” To sum up, the coffers of the 
penurious men are moldy and the keys rust; they wear cloaks which 
hardly reach the thigh; a very little oil-bottle supplies them for anoint- 
ing; they have hair cut short and do not put on their shoes until 
midday; and when they take their cloak to the fuller they urge him 
to use plenty of earth so that it will not be spotted so soon. 

In all ages and in all lands there have been penurious men. The 
skill of Theophrastus consists in his selection of such universal types, 
in his choice of illustrative incident, and in his economy of expression. 
By the addition of one bit of penurious behavior to another he has 
exemplified the dominant trait and has at least by implication given us 

4 But the dates of his life are given by modern authorities as 372-287 b.c. 

5 8 


a theory of the structure of personality. It is clear that to him penuri- 
ousness as a personal trait is a dynamic and directive force, a true 
mainspring of conduct. It is stable, predictable, self-consistent,, and 
compulsive. In the case of the penurious man nineteen samples of 
conduct exemplify the inner trait; they seem to issue directly and 
unambiguously as motor manifestations of inner disposition; they are 
not at all “specific habits” unrelated to one another, but form a 
coherent and meaningful style of life, which to be understood must 
be referred to the central trait. 

Whether he is in business or at home, whether he is entertaining 
his friends or making his toilet, whether he deals with his wife, his 
servants, his friends, his debtors, his neighbors, or his deity, the man 
is dominated by his penurious nature. His conduct does not “depend 
upon the situation” nor has he “as many social selves as there are 
individuals who recognize him.” He is consistently parsimonious. It 
is safe to predict that in new situations he would react in the same 
way. Note well: The man did not have merely a ‘habit” of wearing 
his tunic short, and another habit of examining the boundary marks 
of his property, and another of going to market without spending 
any of his money. Each activity is merely one manifestation of a 
central motive. 

The persistence of the theory that a personality is known best 
through its dominant trait is found in the long line of followers of 
Theophrastus. A list of the more eminent Character writers would 
include many names famous in the history of literature: John Earle, 
Samuel Butler, Ben Jonson, John Donne, Richard Steele, Joseph 
Addison, Samuel Johnson, Jean de la Bruy ere, the Marquis de Vau- 
venargues, and George Eliot. In America Character writing has been 
recently resumed. By entitling his book of character-sketches En- 
cmistics (1926), Stark Young has made an admirable and deliberate 
translation of Theophrastus’ XagaxTf|(>. 

Although Jean de la Bruyere (1645-1696) confined himself to 
Portraits, they were so universalized that they have taken their place 
with the very greatest of Character writing. Issued originally as satire 
of the manners and personalities of his times, these sketches are ap- 
plicable to human beings in any age or land. An illustration (from 
Aldington’s translation) will show how much less formal was his 
writing than that of Theophrastus, and how he contrives his effects 
without explicitly naming a dominant trait. 




Giton has a fresh complexion, a full face and bulging cheeks, a 
fixed and assured gaze, broad shoulders, a projecting stomach, a firm 
and deliberate tread. He speaks with confidence; he makes those who 
converse with him repeat what they have said and he only moderately 
enjoys what is said. He unfolds an ample handkerchief and blows his 
nose noisily; he spits to a great distance and sneezes very loudly. He 
sleeps by day, he sleeps by night; he snores in company. At table and 
in walking he occupies more room than anyone else. He takes the 
center and walks with his equals; he stops and they stop; he walks on 
and they walk on; all regulate themselves by him; he is not inter- 
rupted, he is listened to as long as he likes to talk; his opinion is ac- 
cepted, the rumors he spreads are believed. If he sits down you will 
see him settle into an armchair, cross his legs, frown, pull his hat over 
his eyes and see no one, or lift it up again and show his brow from 
pride and audacity. He is cheerful, a hearty laugher, impatient, pre- 
sumptions, quick to anger, irreligious, politic, mysterious about cur- 
rent affairs; he believes he has talents and wit. He is rich. 

In this sketch it is difficult to name the central trait in Giton’s 
nature, if indeed only one trait is indicated, but it is not difficult to 
comprehend the man and to recognize him among our own acquaint- 
ances. A peculiar style of life permeates Giton’s every activity. Cer- 
tainly he is extroverted, expansive, ascendant, conceited, selfish, and 
socially insensitive; but these traits are not separate and distinct 
springs of action; they flow together in a style of expression. La 
Bruyere does not dissect his character but holds the mirror for his 
full reflection. 

There are other features of this sketch of interest to modem psy- 
chology. Giton has a physique that somehow seems thoroughly con- 
gruous with his temperament ; 5 he also enjoys economic security 
which enhances his self-assurance. By giving the reader these two 
bits of “explanatory” (genetic) information the author enhances 
enormously the reader’s understanding of Giton’s personality. The- 
ophrastus described the ruling passion. La Bruyere by adding a 
small amount of genotypical data draws a fuller portrait. Another 
feature of interest is the selective artistry of the author. Undoubtedly 
Giton had many contradictory aspects of nature, sets of interests 
and forms of conduct not here disclosed. All types , however, must 

5 The significance of the pyknosomic physique is discussed on pp. 73 f. 

6 o 


be simplifications; so too must all Portraits if they are to be symbolic 
of an entire range of personalities. A full-length picture would in- 
evitably introduce elements of uniqueness and would sacrifice its uni- 
versality in the interest of the individual case. 

Character writing, then, for all its virtues of brevity and com- 
pactness, and for all the beauty of its prose sonnets, is a limited 
medium. At its best it produces stylized and simplified depictions of 
human nature, universal in their significance but for that very reason 
remote from the vital individuality of living people. Few character 
writers, it should be added, follow Theophrastus and La Bruyere in 
keeping the psychological interest uppermost; the majority drift off 
into the by-paths of burlesque, preciosity, or moral censure. 

Still greater treasure for the psychologist lies in the world’s store 
of drama, biographies, poetry, and fiction. After reading the magnifi- 
cent studies of personality written by literary genius the psycholo- 
gist feels ineffectual and a bit foolish. He may even agree with Zweig 
that writers like Stendhal, Amiel, Tolstoy, Carlyle and Proust are 
“giants in observation and literature, whereas in psychology the field 
of personality is worked by lesser men, mere flies, who have the safe 
anchorage of a frame of science in which to place their petty plati- 
tudes or minor heresies.” 6 

Since literature and psychology provide the two most important 
methods for the study of personality the comparison between them 
must be pursued somewhat further. It may be true, as Zweig alleges, 
that the psychologist is heavy-handed and unimaginative, whereas the 
literary artist captures the subtleties of personality with precision and 
delicacy. It may even be true that whenever the psychologist suc- 
ceeds in spite of his scientific cliches in representing personality with 
fidelity, he seems then to be saying only what some literary genius 
has already said more agreeably and artfully. Psychology would seem 
to come off very badly in the comparison, for only a pedant could 
prefer his own scientific analysis to the glorious characterizations of 
Dickens or Ibsen. Yet on the other hand, only an esthetic bigot can 
deny that there is a place for some scientific complement to the un- 
disciplined artistry of literature in the attempt to understand, and 
even to depict, personality. 

8 Stefan Zweig: Adepts in Self Portraiture: Casanova, Stendhal, Tolstoy. Trans. 
1928. For another comparison somewhat more respectful toward the attainments of 
psychology see Max Eastman, The Literary Mind: Its Place in an Age of Science , 


6 1 

The subject matter of literature is entirely idiographic. It is the 
single person, the “particular truth” that stands revealed. Whatever 
broader applications literature may have are merely implicit and 
are usually debatable. The psychologist, on the other hand, has an 
inescapable interest in the discovery of general principles, of laws 
of human behavior. In recent times, as Chapter I explained, some 
psychologists have grown dissatisfied with the exclusively nomothetic 
outlook of their science, and have approached more closely the prob- 
lems of individuality hitherto consigned to literature. But there re- 
mains, however, a difference: the literary writer cares primarily for 
the individual case, leaving to the reader the task of generalizing the 
insight he gains. The psychologist, while studying the single case, is 
never content until he himself has made appropriate generalizations. 
The generalizations are not, or should not be, concerned only with 
the operations of an hypothetical “average” mind. Their aim is rather 
to state explicitly the principles by virtue of which unique personali- 
ties are created by nature and understood by men. 

Another important difference exists in the context in which per- 
sonality is studied. The literary investigator may give the full social 
setting and portray the cross-currents in which his character de- 
velops, confining himself to complex levels of personality. He de- 
velops his character in the stream of life. The psychologist likes to 
disregard the complexities of the social setting, to fasten the person- 
ality as-it-is for analysis, to reduce so far as possible the confusion 
of surrounding variables, and to seek the elementary features of con- 
duct which can be separately studied. Lie prefers the laboratory or 
the clinic to the stream of life. Here, of course, lies a serious danger. 
Too much analysis and control may destroy his datum. But with 
due care his method of isolation brings many significant rewards. 

The effects of literature are gained by skillful selection and exag- 
geration. The petits faits vrais are made prominent, currents unsuited 
to the interests of the author are omitted. And yet the very incidents 
that heighten the character may to a degree falsify it. This distortion 
is common in both fiction and biography. The psychologist, how- 
ever, is permitted no artistic accentuation, and illustrative incident 
must not be used until it is known to be both recurrent and diag- 
nostic. Exaggeration apropos is not allowed. Also, in the arbitrary 
scope of a work of literature a lack of discipline not permitted 
science is shown. The telling may begin anywhere, and leave off 
abruptly; it may be simplified as much as the author desires to secure 



an eifect. The psychologist must give reasons for his beginnings and 
his endings, as well as for his inclusion and his exclusion. 

The artist strives to be entertaining and engaging, to communi- 
cate his own images, to express his own biases. One measure of his 
success is the responsiveness of his readers. The psychologist is per- 
mitted only to discover and record. His primary aim may be to in- 
struct but never to entertain. His work must exclude his own bias, 
be true to the scientist’s conception of fact, and his success is meas- 
ured by sterner criteria than the reader’s applause. 

In gathering material, the writer draws from his casual observa- 
tion of life, the psychologist from controlled investigations. The 
writer may present his observations in epigram and in pretty phrases, 
the psychologist must use exact and standard terminology. In litera- 
ture rapid and bold inferences may be made, in psychology the in- 
ferences requiring proof step by step are slowly and cautiously 
drawn. It is unnecessary for a literary work to be “proved” or sub- 
mitted to the test of repetition; it is even unnecessary for the author 
to be completely consistent in his own statements. He need not fit his 
observations into a conceptual system for the purpose of testing a 
general theory. All these requirements bind the psychologist. “The 
scientist must submit to the judgment of others, not merely the con- 
clusions which he reaches, but also the premises with which he began 
and the methods which he uses in developing his work from them; 
but the artist, within very wide limits, at least, is allowed to choose 
whatever premises he likes, and as a result, it is by no means necessary 
that all good artists should agree to anything like the same extent that 
all good scientists must agree.” 7 

Freed from the bondage of scientific terms, the writer can speak 
recklessly of the course of nature. He can ascribe causes and assign 
correlates at random. One of his characters may have “menial blood 
in his veins,” another a “weak chin.” A hand may possess “a wonder- 
fully cruel greed” and a blond head “radiate fickleness.” Such un- 
disciplined metaphors give cadence and inspire a kind of bland 
credulity, but for science they are mere idle phrases. Recently a 
famous professor of literature in describing a character wrote, “The 
nose, almost invariably the index of mental power, was perfect in 
fullness, straightness, and strength.” No psychologist could write 
such a passage without being torn limb from limb by his professional 
colleagues! iTyTY 'k; : 

7 Joseph W. Krutch, The Modern Temper, Harcourt, Brace, 1929, p. 152. 



Psychology will not supplant literature, nor will the hauteur of 
artists hinder the growth of psychology. The two methods are dis- 
tinct and complementary. If psychology today is discovering only 
what literature “has always said” it is nevertheless giving precision 
and general application to the ancient truths. Less enjoyable, it is 
more disciplined; less subtle, it is more verifiable; less artful, it is more 


Even more ancient than Character writing and closer in spirit to 
modem science is the doctrine of the humors and their correspond- 
ing temperaments. This is the oldest characterological theory of 
which there is any record. Growing out of the four-part cosmogeny 
of Empedocles it has had an almost unbroken history. 8 With rela- 
tively few changes it has endured from the dawn of history down 
to the latest text-book in psychology. 

In the following table the development of the doctrine in classic 
times is reviewed. Its original logic rested upon the belief that man is 
a microcosmic reflection of nature. He should therefore in his own 
being express all the properties of the cosmos. 

Cosmic Elements 

Their Properties 




T emper aments 



cir . 450 B.C. 

cir . 400 B.C. 


warm and moist i 


j ■■ 



cold and dry 

Black Bile 



warm and dry 

Yellow Bile 


Water ' 

cold and moist 



The classical doctrine ascribed peculiarities of temperament to the 
“humors” of the body. In the light of modern physiology and endo- 
crinology the list of specific “humors” advanced by Hippocrates has 
been completely abandoned, but the principle of psychophysical cor- 

8 Taken together the following references reconstruct the whole interesting 
story: Ben Jonson, Every Man Out of His Humor; V. Laehr, Literatur der Psychia- 
tric, Neurologic und Psych ologie von 1459-1-1 $9, 3 Vols., 1900; A. A. Robaclt, A 
Bibliography of Character and Personality, 1927; L. Klages, The Science of Charac- 
ter, pp. 144-149; W. B. Pillsbury, The History of Psychology, 1929, 37-44; P* 
Malapert, he Caractere, 1902; A. A. Roback, The Psychology of Character, chap, iii 
and Ft. II; W. Stem, Differ entielle Psychologic , Appendix L 



respondence remains. Chemical substances, notably the hormones, are 
now known to affect the working of the nervous system in ways that 
were only dimly surmised by the ancients. Modem science has shown 
these substances to be even more powerful, more numerous, and more 
varied in their influence than Hippocrates supposed. 

Later variations of the theory took the form of subdividing or 
renaming the temperaments, and of modernizing the conception of 
humors. The doctrine profoundly influenced medicine, especially 
down to the time of Harvey’s discovery of the circulation of the 
blood. It likewise influenced literature and art. It was reworked by 
Kant, Wundt, Ribot, Fouillee, Ribery, Azam, Malapert, Paulhan, by 
Hoff ding, Bahnsen, Herbart, Kiilpe, Ebbinghaus, Meumann, Spurz- 
heim and Klages. It has claimed the attention of intellectual giants as 
well as charlatans, and still survives in a modified form in our new era 
of research. 

There are two principal reasons why interest has been so per- 
sistent in this venerable approach to temperament. In the first place, 
the happy guess that temperament, the emotional groundwork of per- 
sonality, is conditioned above all else by body-chemistry has been in- 
creasingly borne out in modern research. 9 In the second place, the 
four-fold classification of temperament is still useful, because of cer- 
tain fundamental dimensions of emotional response that it implies, a 
fact that Wundt, of all the writers on the subject, seems most clearly 
to have perceived. 10 According to Wundt men may differ in the 
characteristic speed of emotional arousal, or in the characteristic in- 
tensity of the response. The four temperaments are essentially the re- 
sulting combinations in a two dimensional scale of emotionality. 









The definition of temperament offered on page 54 recognizes as 
important distinguishing factors in temperament these Wundtian di- 
mensions of strength and speed. But breadth and depth are likewise 

9 A readable presentation of the case is R. G. Hoskins, The Tides of Life, 1933. 

10 W. Wundt, Grundziige der physiologischen Psychologie, 5th ed., 1903, 
Vol. III. 


important variables, and it so happens that the classical types may be 
ordered under these dimensions as well. 









Or still another dimensional arrangement may be made in terms 
of predominant affective tone, pleasantness and unpleasantness, paired 
with the kinetic dimensions of excitement and calm. 

i Pleasant 








In short, because of their elasticity the temperaments fit into vari- 
ous dimensional or quantitative schema, thereby satisfying the re- 
quirements of various investigators. Originally, the temperaments had 
merely a qualitative coloring; the choleric being irascible, the san- 
guine, alertly hopeful, the melancholic, sad, and the phlegmatic 
merely apathetic. But these colorings, it turns out, fit nicely into a 
number of modern dichotomous schemes. 

The longevity of the theory is thus due in part to its flexible 
nature and in part to its implicit recognition of the underlying effect 
of body-chemistry. In this way the venerable conception of tempera- 
ment still serves a purpose. But the gain is not great; sounder and 
newer formulations are now needed. 


The art of discovering characteristics of personality from the out- 
ward appearance, and especially from the configuration, cast, or ex- 
pression of the face, is called physiognomy . Probably mankind has 
always practiced this art. The oldest treatise on the subject is one 
entitled Physiognomonica attributed, probably incorrectly, to Aris- 



rode . 11 Ancient though it is it refers to three methods each having its 
special adherents in the still more distant past. The first method 
sought resemblances in appearance between men and animals and 
assumed that when the form of a man is reminiscent of that of a 
particular animal he must be endowed with some of the same psychic 
qualities. The man who looks like a fox must be sly. The second 
method was a kind of racial typology. The man who is excessively 
pale or dark should be judged a coward, for— the author asks— are 
not Ethiopians dark and women pale, and both alike cowardly? The 
third method took as its basis the facial expressions engendered by 
emotion, and sought in the face muscular traces of angry, fearful, 
or lustful habits of thought and expression. 

The author of this ancient treatise gives a list, “a complete list,” 
of the sources from which physiognomic signs are drawn. They are: 
“movements, gestures of the body, color, characteristic facial expres- 
sion, the growth of the hair, the smoothness of the skin, the voice, 
condition of the flesh, the parts of the body, and the build of the 
body as a whole.” tie adds, “But inferences drawn from the parts of 
the body are less secure than those based on facial expression of 
character and movements and gesture. In general it is silly to rely 
on a single sign: you will have more reason for confidence in your 
conclusions when you find several signs all pointing one way.” From 
this treatise several valuable suggestions emerge, especially the re- 
quirement that signs must be consistent with one another if they are 
to sustain a judgment, and the suggestion that signs which spring 
from active habits of expression are more to be trusted than simple 
structural characteristics which are innate and unchanging. The 
author is keen likewise in his observation that certain states of mind 
have no recognizable bodily counterparts. The specific content of a 
man’s knowledge or belief, for example, are not physiognomically 

With the Aristotelian revival in the thirteenth century physiog- 
nomy became once more popular, and has in fact had an unbroken 
history since that time. The quantity of literature on the subject is 
far more striking than its quality, for the practical nature of its appeal 
brought it early under the patronage of quacks and charlatans, where 
unfortunately it has largely remained. So great were its abuses in the 

11 This treatise which covers only 24 pages is translated by T. Loveday and 
E. $. Forster and appears in the Qpnscula> Vol. 6 ? of the Works of Aristotle (ed. by 
W. D. Ross), rqrv 



eighteenth century that George II by an act of parliament deemed 
all persons pretending to have skill in physiognomy rogues and vaga- 
bonds, to be publicly whipped or sent to houses of correction. A 
similar law had been passed in the sixteenth century in the reign of 
Queen Elizabeth. 12 

The crusade against fraudulent physiognomists is still being 
waged, not so much by law as by professional psychologists. It may 
seem strange that the “impractical” psychologist should be forced 
to protect “hard-headed” business men from these charlatans, but 
such is the case. Dunlap has given four reasons why popular systems 
based on physiognomy appear to succeed. 13 The first is that the 
actual value of the readings is rarely checked; the second, that a 
few, though not many, of the physiognomists make surprisingly good 
guesses based on indications which they do not really understand, but 
falsely ascribe to their own systems. 14 An additional reason is found 
in the clever salesmanship of the analysts (to which even business- 
men are not immune) . Finally, the “experting” is done at the instance 
of some director who soon loses interest, and pigeonholes the results. 
But the physiognomist goes on to the next job with a lordly fee in his 
pocket, and impresses his new clients with his reputation for having 
successfully served the first corporation. So the cycle continues. 

The shadowy history of systems of physiognomy may be dis- 
regarded in favor of the challenging fact that men have always found 
some indirect assistance in their judgments of others through the 
observation of their physical expressions: their eyes, their cast of 
countenance, their facial play, their posture, build, and manner. Even 
excluding all actual movement, the body at rest retains traces of its 
habits of exercise, the face seems to betray one’s way of life. There 
is also theoretical justification for judgments based on physiognomy: 
growth is largely regulated by the glands of internal secretion, so too 
is the emotional life. Physical features therefore may logically be ex- 
pected to reveal peculiarities of temperament. 

In Figure 9 are presented certain traditional drawings of the 
facial types supposed to correspond to the four temperaments. (The 

12 Encyclopedia Britannica , nth ed., VoL 21, p. 550. 

13 K. Dunlap, “Reading of Character from External Signs,” Set. Mo., 1922, 15, 2, 

14 F. Gall, the phrenologist, probably gave the correct interpretation of this 
point when he asserted that the apparent success of practicing physiognomists came 
from their judgments of muscular expression, posture, and movement, and not from 
the cast of the facial features, F, Gall, On the Functions of the Brain , 1835, Amer. 
trans., VoL V, p, 2 66. 



Figure 9 

Physiognomic Representations of the Four Temperaments 

familiar with the characteristics of the four temperaments can recog- 
nize them in these drawings. Shown to several hundred judges, the 
percentage of correct designations was as follows: 



No. 1. Melancholic 83% 

No. 2. Choleric 86% 

No. 3. Phlegmatic 81% 

No. 4. Sanguine 80% 

The commonest errors are the confusions between the two intense 
temperaments (melancholic and choleric) and between the two un- 
emotional temperaments (phlegmatic and sanguine). It may rightly 
be objected that both the pictures and the temperaments are pre- 
sented in their extreme forms, and that moderate or mixed instances 
would certainly not be so easily recognized. It is also true that stereo- 
types and verbal habits aid in correct judgments. Number 1 looks 
like the traditional love-sick poet who is by reputation melancholic. 
Similar associations may be recorded for the others. Number 2 ap- 
pears to be a “fighter” clearly endowed with choler ; Number 3 is 
sleepy, flaccid, and phlegmatic; Number 4, because of the suggestion 
of vacuous optimism, is judged sanguine. But these inferential and 
associative judgments are ultimately physiognomic, for whence came 
our stereotypes of the melancholy poet, the fighter, the sluggish 
beef-eater, and the shallow optimist if not through actual experience 
of just such correlations between physical and temperamental charac- 

There are two entirely different types of physiognomic diagnosis. 
There is first that based on bony structure, and second that based on 
muscular set. Since bony structure cannot be changed by experience, 
it would not seem to be a promising index to personality, for person- 
ality is largely a product of education, experience, conflict, and 
adaptation to the environment. Heredity alone gives us our bony 
structure; but it alone does not give us our personalities. Therefore, 
the correlate of bony structure or constitutional build is tempera- 
ment, the native factor in personality. Muscular set, the second cri- 
terion, is the agency of movement and is in turn influenced by habits 
of movement. The professional work of the sailor, the blacksmith, 
the clergyman, the teacher and the prize-fighter create diverse pat- 
terns of muscular strength and flaccidity, specialization and exercise. 
It is from these muscular sets, therefore, that we seek information 
regarding the acquired factors in personality, the products of experi- 
ence. Needless to say in the physiognomic judgments of everyday 
life we do not distinguish the separate diagnostic significance of the 



bony structure and of the musculature, but this distinction is neces- 
sary as a first step in physiognomic theory. 

'The drawings in Figure 9 depend for their effect upon both the 
bony and the muscular structure. Considering first the muscular 
structure, it may be pointed out that the crucial facial lines in Num- 
ber 1 are the vertical furrows in the brow, and the downcast eyes. 
The former feature is present characteristically in states of un- 
pleasantness and during thought. Both features would be expected 
in a temperament which is slow to be excited, but intense, when 
aroused. 35 In Number 2, the choleric nature is betrayed both by the 
heavy seaming of the face, the result of frequent emotional seizure, 
and by the forward direction of the eyes, displaying the attentiveness 
required for prompt reaction. In Number 3, the facial lines do not 
follow the muscular divisions of the face; they are due to fat rather 
than to feeling. The eyes are inattentive, and suggest the slowness 
which is characteristic of the phlegmatic temperament. Number 4 
has a smooth face, devoid of muscular traces of intense emotional 
experience, and wide open eyes, indicating a readiness for quick re- 
sponse; these are the two distinguishing features of the sanguine 
person. The analysis thus far is strictly in the psychological tradition 
of Piderit, Wundt, Boring and Titchener. 16 Popular physiognomists 
would undoubtedly wish to add more detailed readings based on the 
height of the brow, shape of the ears, contour of the nose, color of 
the eyes or hair, but so far as is known these features yield abso- 
lutely no reliable positive correlations with characteristics of person- 
ality. 17 

Personality is only one of four determinants of facial expression, 
and it is by no means easy to “see through” the effects of the other 
three influences and find the reflection of the true personality. First 
of all there is the limitation placed by constitutional build. Faces may 
be short or broad, with thick lips or thin, with deep-set or protruding 
eyes. The play of expression is constrained by such native structures 

15 There is some evidence in favor of the view that pleasant emotions are quick 
to arouse, while the unpleasant are slower to arouse and perhaps more enduring. 
F. H. Allport, Social Psychology, pp. 85-94. This fact would explain why the slow 
melancholic temperament should suffer from a characteristically negative (un- 
pleasant) feeling tone as well as from latency and perseveration of response. 

16 Cf. E. G. Boring and E. B. Titchener, “A Model for the Demonstration of 
Facial Expression,” Amer. J. Psychol., 1923, 34, 471-485. 

17 Cf. Cleeton and Knight, “Validity of Character Judgments Based on External 
Criteria.” /. Appl. Psychol., 1924, 8, 215-231. Also Paterson and Ludgate, “Blonde 
and Brunette Traits: a quantitative study,” J. Pers. Res., 1922, 1, 122-128. 


7 1 

of the countenance. Although, as will soon be shown, the native 
structure itself may have some slight significance for personality (as 
a means of indicating racial or endocrine types), it is at the same 
time often a misleading mask. A face, for instance, with coarse or re- 
pulsive features may blind the observer to the play of wistful or 
friendly expressions. 

Another “non-personal” determinant of facial expression is pro- 
vided by the primary emotions, such as fear, anger, amusement, dis- 
gust, universal among men, and essentially the same from country to 
country and from age to age. This common instinctive groundwork 
underlying facial expression is not related to individuality but results 
from a common biological constitution. Consistent with their pre- 
occupation with mind-in-general, it is in this field that psychologists 
have expended most of their efforts . 18 

A third interlocking influence is the presence of conventional 
(racial and regional) standards of expression which individuals (even 
children) adopt, and through which they actually come to resemble 
their fellows . 10 It has been frecjuently noted that to the occidental 
eye, Orientals “all look alike,” and conversely, incredible though it 
may seem to us, Orientals have complained that Americans are hard 
to differentiate by their faces. These racial similarities are due in part 
to constitutional type but also in part to the assumption of standard 
expressions. It is usually difficult for the foreigner to recognize the 
play of individuality within the racial pattern. 

There are also customs which require affected facial expression. 
The tired hostess, whatever her inner conflict, must assume the 
“awfully glad” look when her guests arrive. A gentleman, according 

18 For instance, there is plentiful research bearing upon the ability of judges to 
recognize common and basic emotional expressions in the face. This work derives 
primarily from Darwin whose interest, of course, was confined to the, serviceable 
vestigial habits common to all men and recognizable in all races and in all indi- 
viduals. Actually, however, experiments show a surprisingly low degree of success 
in identifying emotional expressions unless the judge is aided by knowledge of 
context and stimulus. Landis helps to explain these poor results by demonstrating 
that there is after all the important variable of individuality . For any single person 
the manner of expressing, say, disgust or mild amusement, is quite constant, but 
from individual to individual there is marked variation (C. Landis, /. Comp . 
Psychol 1924, 4, 447-509). To be sure, in states of intense emotional involvement 
the expression conforms more closely to the norm for the human race. Emergencies 
always have a way of leveling personal forms of expression to a form common to the 

19 The role of this unconscious imitation of expression was recognized by the 
philosopher, Immanuel Kant, who pointed to the curious fact that congenial married 
couples grow to have a similar cast of features .Anthropologies Part I, sec. 52. 



to Lord Chesterfield, must “never show anger.” In the ancient cul- 
tures of the Orient, even more severe masking is required. Lafcadio 
Hearn speaks somewhere of his amazement at seeing the tortured 
and tragic expression on the usually placid face of his Japanese valet 
when the latter thought he was unobserved. All of our various social 
masks (of “interest,” “friendliness,” and “equanimity”) belong to 
the conventionalized group of expressions whose purpose seems to be 
to smooth the path of social intercourse by submerging the often 
turbulent life beneath. It requires a shrewd observer to distinguish 
in facial expression the difference between assumed interest and real 
absorption, between authentic self-assurance and the pose of bravado. 

But behind the universal, instinctive expressions common to all 
mankind, and behind racial and conventional standards, there exist 
true differences in facial lines and casts of countenance that can be 
attributed only to individual habits of thought and emotion. In order 
to practice the arts of physiognomy one must find means, if one can, 
of distinguishing the effects of each of these influences in order to 
read through the “outer” determinants to the “inner.” At present 
the situation is simply this: we know that facial cues are revelatory 
of “the life within” but we find these cues obscured by the influence 
of such “non-personal” accidents as membership in a common species, 
hereditary structure, race, and convention. 

Physiognomy, as has been pointed out, deals not only with the 
problem of muscular expression, which has just been discussed, but 
also with the almost wholly independent problems of bony structure 
and constitutional habitus. It is only in recent years that psycholo- 
gists have interested themselves in this second class of problems. In 
novels and dramas, various psychophysical relationships have long 
been assumed. Julius Caesar says: 

Let me have men about me that are fat; 

Sleek-headed men, and such as sleep o’ nights: 

Yon Cassius has a lean and hungry look; 

He thinks too much: such men are dangerous. 

Julius Caesar, Act I, sc. 2. 

It was not Shakespeare, however, who inspired the psychological 
laboratories, clinics, and other psychometric centers to work on this 
problem. The credit goes to a German psychiatrist, Ernst Kretsch- 
mer, whose brilliant book Physique and Character, published first in 



German in 1921, provoked a flood of investigations, which have con- 
firmed in part his striking claims. 

Kretschmer presented evidence that in mental hospitals elongated 
and frail “asthenic” physiques were found most frequently among 
dementia praecox patients, and that short and rounded “pyknic” 
physiques were most frequent among manic-depressive patients. Re- 
duced to its barest statement, this was Kretschmer’s finding. 

Dissatisfied, however, with so bald a statement of the correlation, 
Kretschmer elaborated this single finding into a theory of the re- 
lationship between physique and normal personality, proceeding from 
the common premise that the abnormal is merely an exaggeration of 
the normal (cf. pp. 13, 76). Thus elaborated the theory requires that 
tall, slender physiques wherever found be ordinarily associated with 
such schizothymic qualities as introversion, formalism, idealism and 
romanticism (“normal” variants on the syndrome of dementia 
praecox). Correspondingly, physiques which are rounded, heavier, 
shorter, with larger body cavities should belong to individuals who 
in the main are cyclothymic; sometimes moody, but often jovial, 
and predominantly extroverted, realistic, and objective (reflections 
of the manic-depressive make-up). 20 Kretschmer distinguishes two 
additional types of body-build, the athletic and the dysplastic, both 
of which, like the asthenic, are supposed to be associated with the 
introverted pattern of personality. 

A more detailed exposition of Kretschmer’s contentions and of 
the investigations which they have provoked would be beyond the 
scope of this chapter. 21 Passing directly to a summary evaluation of 
the evidence to date, it is unfortunately necessary to report that the 
exuberance of Kretschmer’s claims and the enthusiasm of his many 
disciples need correction and restraint. Taken at the pathological 
level the evidence is favorable enough, though not as favorable as 

20 In the schematic drawings in Figure 9 the reader has already noted the slender 
and delicate (asthenic) physique of the melancholic temperament (No. 1). The 
physical-mental correlation here indicated by the humoral theory is quite in agree- 
ment with Kretschmer’s doctrine. The hormones are, according to both theories, 
responsible for the fragility of the bodily structure and for the introverted, idealistic, 
withdrawing attitude. Somewhat less clearly perhaps the drawing of the phlegmatic 
type (No. 3) depicts the pyknic physique, and the traditional conception of this 
temperament is* not incompatible with Kretschmer’s picture of the extroversion and 
realism found in this physical type. 

21 Reviews of the literature on this subject may be found in L. Polen, “Korper- 
bau und Charakter ” Arch. f. d. ges. Psychol 1928, 66, 1-116 (a complete bibliog- 
raphy up to 1928); O. Klineberg, Race Differences, Harper, 1935, pp. 56-66, 


Kretschmer claims . 22 At the normal level there are many totally nega- 

tive investigations. 

The good and the bad points of this type of modem physiognomy 
can be best understood, not by citing the conflicting evidence, but 
by examining the presuppositions of the work. Kretschmer’s original 
hypothesis, it will be shown, is in part unsound, and therefore should 
not be expected to yield uniformly positive results. 

In the first place, Kretschmer’s theory requires that “character” 
be innately determined. But is it? Is it not rather the temperament 
(the emotional ground upon which “character” develops) that is 
closely controlled by the chemistry of the body and therefore asso- 
ciated with physical build? Kretschmer should not have attempted to 
write of “physique and character,” nor of physique and personality, 

22 According to Kretschmer the incidence of physiques among the two classes 
of psychotics in a group of over 4,000 cases in mental hospitals is as follows: 

Circular Insanity Schizophrenia 

Constitimonal Type ( Per cent of cases) ( Per cent of cases) 

Pyknic and pyknoid 66. 7 12.8 

Leptosome (asthenic) and athletic 23.6 66.0 

Dysplastic 0.4 11.3 

Atypical 9.3 9.9 

(From O. Klineberg, op. cit. y p. 69.) 

On the other hand critics have pointed out that age is an uncontrolled influence 
in this table; ( e.g C. R. Garvey, Psychol. Bull., 1933, 30, 567, 739). Schizophrenia 
generally strikes early in life when the body is still slender. The average age of 
manic-depressives is higher, and with age many physiques become pyknic. 

23 One example may be briefly described. P. S. Cabot, employing a group of 
200 boys in late adolescence, for whom anthropometric records of growth were 
available for twelve preceding years, undertook to check on Kretschmer’s claims. 
(“The Relation between Characteristics of Personality and Physique in Adolescents,” 
forthcoming in Genet. Psychol. Monog.) For the physical criterion he employed 
various types of anthropometric indices which seemed fairly to represent the 
asthenic, pyknic, and athletic physiques; he also used ratings of these physiques 
based upon impressions of the boys’ appearance. On the side of personality his 
criteria consisted of dozens of rating scales, tests, and questionnaires all pointed in 
such a way as to obtain the optimum information concerning each boy’s personality 
(with special reference, of course, to Kretschmer’s claims regarding introversion, 
withdrawing, autism, leadership, realism, objectivity, and the like). 

The results of the investigation offer little support for any of Kretschmer’s con- 
tentions. The differences in personality shown by each physical type seemed, on 
the contrary, just about what one would expect according to a wholly different 
theory. Those boys who had solid and substantial physiques (whether ' pyknic or 
athletic) seemed on the whole to develop personalities that were extroverted , out- 
going, realistic, and dominant. In short , a “good” physique disposes the boy to de- 
velop strong traits of self-expression. Social acceptability and physical vitality, or 
else compensation, seem to be the determining factors. These factors, all of "them 
dependent upon the interaction of the boy’s constitutional make-up with environ- 
ment, are so outstanding as to cast serious doubts upon the native predetermination 
of personality required by Kretschmer’s theory. 



but rather of physique and temperament. Had he done so he would 
not have entered the hazardous claim that the finished portraits of a 
personality can be paralleled by corresponding physical types. Deal- 
ing with temperament he would have been on safer ground, for emo- 
tionality and physique must indeed be expected in some basic way to 
correspond (since both are partially the products of the glands of 
internal secretion). There is a marked difference between traits that 
are the product of all formative influences (such as extroversion, 
dominance, leadership, autism, and the like) and the simpler and more 
basic qualities of temperament (cyclothymia, melancholy, euphoria, 
phlegmatism). It is in this second group of features, and not in the 
first, that he should have sought his correlations. 24 

Kretschmer’s strong nativistic bias leads to a further difficulty 
when the question of hereditary mixtures is concerned. Constitu- 
tional habitus is clearly not a Mendelian unit, neither is “character” 
a unit trait. Considering our long lines of mixed ancestry, it is re- 
markable that he should find so many pure types to illustrate his 
theory. 25 Actually, of course, most people are extreme neither in 
physique nor in temperament; the types are mixed. And yet if 
Kretschmer has failed to solve the riddle of the inheritance of tem- 
perament and body-build, so too has every other investigator; one 
cannot be too critical on this account. 

All the inadequacies of the concept of “types” are realized in the 
Kretschmerian scheme as it stands at present (cf. pp. 13-15). For 
example, Kretschmer takes his departure from the two major forms 
of functional mental disturbance (dementia praecox and manic-de- 
pressive psychosis). He finds two corresponding physiques. But what 
would he do if he added to his basic list of “character types” other 
clinical forms of disturbances in personality, for example the epilep- 
toid, the paranoid, or the psychopathic inferior? And why, in his 
theory, should such utterly different physical types as the dysplastic 

24 The only results supporting Kretschmer’s theory have come from mental 
hospitals where one finds the effect of extreme temperamental (emotional) disturb- 
ance upon the psychosis. It is not the psychosis itself, however, that corresponds to 
the constitutional type, but rather the temperament underlying the psychosis. 

25 The averaging out, through mixed inheritance, of all extremes in physical 

build, is, however, to some extent prevented through a curious phenomenon of as- 
sortative mating. C. B. Davenport has shown that slender men tend to marry slender 
women to an extent exceeding chance expectation by 50%; fleshy men and fleshy 
women marry to an extent exceeding chance by 80%. Hence the children of these 
marriages tend, to some extent, to perpetuate the more or less pure asthenic and 
pyknic physiques. P*oc. Assoc . Res. Nerv . & Ment. Dir., 1934, 2I ** 2 7« 



and athletic be lumped together with the contrasting asthenic phy- 
sique, and all three then be expected to parallel the schizothymic 
mental make-up? 

Empirical evidence has thus far failed to prove that Kretschmer’s 
correlation occurs anywhere else than in cases of mental disease. He 
believes, nevertheless, that what holds for the extremes must hold as 
well for the means. But should it? Is the normal personality simply 
an undistinguished edition of the mentally diseased? W e do not hold 
this view in reference to organic conditions. There is no continuum 
of states from cancer to no-cancer. The patient either has a malig- 
nant growth or else he hasn’t; there are no intermediate conditions. 
Similarly a diseased mind is in many respects functionally quite differ- 
ent from (and not merely an exaggeration of) the normal mind. 

The belief in the perfect continuity of the normal and abnormal, 
which most psychiatrists and psychologists share with Kretschmer, 
has resulted in the rapid multiplication of studies of disordered people, 
partly because, confined as they are to institutions, they are easily 
accessible, and partly because the extreme nature of their disorders 
makes them more interesting and more spectacular. Actually the 
number of studies of neurotic and psychotic personalities far exceeds 
the number devoted to normal personality, although, of course, the 
ratio in the world at large is precisely the opposite. The uncritical 
carrying over of the point of view of the mental hospital into the 
world outside has made, as in Kretschmer’s case, for serious one- 
sidedness in the psychological study of normal personality. This 
charge is justified, even though occasionally the discoveries of psy- 
chopathology may be of indirect aid to the psychology of normal- 

Returning to Kretschmer’s theory specifically, we conclude that 
extreme glandular imbalance may affect physique and temperament in 
significant ways, whereas normal conditions of glandular balance 
leave a greater play for environmental and experiential determinants. 
What slight effect constitutional type may have in the development 
of the personality of normal people seems to be totally overlaid by 
the more important determinants of education and social experience. 
In cases of abnormality the imbalance is so extreme as to dominate 
the picture. \ ' 

Before summarizing this somewhat lengthy discussion of physiog- 
nomy, we should pay our respects to the most famous physiog- 



nomist of all times, Johann Kaspar Lavater (1741-1801). 26 Many de- 
tails of his system are not helpful, for in some respects they resemble 
the contentions of modem “quack” physiognomists; certainly his 
tendency to combine his readings of personality with moral preach- 
ment may be overlooked. 27 It is Lavater, however, who expressed 
most clearly the exceedingly important hypothesis that all of the 
features of the body are ultimately congruent and consistent. Their 
forms of expression all proceed from a central, unified personality, 
and therefore they must be harmonious among themselves, and must 
all betray the organization of personality within. Laughing eyes, he 
says, do not occur without a laughing mouth, nor, if we are shrewd 
enough to see it, without corresponding “mirthful” qualities in the 
gait, handwriting, and postures of the body. He is emphatic in this 
view: “Everything in man is progressive; everything is congenial. 
Form, stature, complexion, hair, skin, veins, nerves, bones, voice, 
walk, manner, style, passion, love, hatred: one and the same spirit 
is manifest in all.” This contention has recently provided a problem 
for laboratory research, and its approximate correctness in the field 
of expression has been experimentally demonstrated. 23 What is espe- 
cially important in Lavater’s system, then, is his emphasis upon the 
radical consistency of personality both in its inward aspects and in 
its expression. 

In conclusion, physiognomy, although it has had a checkered and 
in part a disreputable career, is an important branch of the psychology 
of personality. But scientific knowledge lags behind naive belief and 
credulous practice. Only a few fairly certain principles can as yet 
be entertained: 

1 , The bony structure of the body (constitutional build) is re- 
lated to personality through the medium of temperament, for 
both the physical habitus and temperament are products of 
bodily chemistry. 

26 His chief work, Physiognomische Fragments zur Beforderung der Menschen - 
kenntnlss und Menschenltebe , issued in 1783-87 in three volumes, has many times 
been republished. Still earlier editions of this work appeared in 1772 and 1775 under 
different titles. A good secondary account will be found in E, Utitz, Charakterologie 7 
pp. 55-60. 

27 Goethe in both his Gespr'dche mit Eckermann and in Dichtung und Wahrheit 
describes vividly Lavater’s evangelical manner of practicing his art upon street 
crowds and in churches. 

28 g. W. Allport and P. E. Vernon, Studies in Expressive Movement , 1933. The 
matter is discussed further in Chapter XVII of the present volume. 

7 8 


2. In cases of extreme glandular imbalance there are likely to be 
severe upsets of emotion, and corresponding changes in per- 
sonality. But “thyroid” personalities or “enuchoid” personali- 
ties occur only in serious cases of glandular dysfunction. 

3 . Within the normal range of glandular functioning (and bodily 
build) other causative influences are vastly more important 
factors in determining the development of personality than is 
the “constitutional type.” 

4. Within this normal range, physical build is associated only in- 
directly with personality. Strong bodies, well-formed, and 
socially approved, predispose people (especially in youth) to 
develop extroverted, realistic, sociable traits; conversely, frail, 
malformed, or markedly atypical physiques tend (in response 
to social and environmental standards) to produce introverted, 
intellectual, or autistic personalities. This finding takes care of 
much of Kretschmer’s evidence, but offers a totally different 
theory (one that is environmentalists rather than nativistic) 
to account for the association of physique and personality 
'within the normal range. 

5. The musculature of the body reflects better than its bony 
structure the influence of life-experiences, and hence from 
facial and postural sets more certain inferences concerning 
personality can be made than from constitutional build (for 
example, a highly educated person or a degenerate can more 
readily be identified by muscular sets than by constitutional 
build) . 

6. In making inferences from muscular set and facial expression 
it is necessary to “read through” such non-personal determi- 
nants as racial membership, local customs, and the universal 
patterns of instinctive emotional expression. Personality is only 
one of the factors that affect the cast of features, or posture 
and movement. 

7. There is considerable consistency among the expressive features 
of the body. 

These conclusions are taken up again and extended in Chapter 





Unlike the preceding three schools of characterology, phrenology 
is exclusively a modern doctrine; it is less than a century and a half 
since it was first promulgated by Franz Joseph Gall (1758-1 828). 29 
Even more than physiognomy it has had its disreputable side, as a 
lucrative “racket” for vainglorious mountebanks. Although its influ- 
ence, enormous in the past, now is waning, phrenology still has its 
devotees. Even today there are phrenological practitioners and maga- 
zines that thrive by simplifying and perverting some of the older 
teachings, none too sound in themselves, by infusing them with the 
popular idols of success, wealth, and fascination. But it is not with this 
slum of psychology that we need concern ourselves. Phrenology has 
another, less familiar, but more significant side. 

The story of the origin, premises, growth and decline of phre- 
nology has many times been told. 30 It need not be repeated here. Nor 
is it necessary any longer to take time to refute the specific assump- 
tions upon which Gall founded his eccentric system: their weakness 
is entirely transparent to the modern psychological reader. 31 But even 

29 The popularity of phrenology immediately commenced with Gall’s first lec- 
tures in 1796. It spread rapidly through the enterprising salesmanship of Gall’s as- 
sistant, Johann Gaspar Spurzheim (1776-1832), who, however, separated from his 
master in 1815, and wrote many independent works, lecturing widely in England and 
on the continent. Spurzheim came to America in 1832, lectured to large audiences, 
was well received at Harvard, and died in Boston. 

Gall’s chief work appeared in four volumes between 1810 and 1819, and was 
entitled, Anatomie et physiologie du systeme nerveux en general , et du cerveau en 
particulier , avec observations sur la possibilite de reconnoitre plusieurs dispositions 
intellectuelles et morales de Vhomme et des anhnaux par la configuration de leur 
tetes. In 1822-25 Gall published a revised edition in six volumes under the title, Sur 
les fonctions du cerveau , etc. The six volume American edition, On the Functions of 
the Brain } etc., was published in 1835. 

The name “phrenology” was invented by Dr. Thomas Forster and was adopted 
by Spurzheim in 1815. Gail himself never used the term, but referred to his doc- 
trine as “organology,” “cranioscopy,” or more often as “the physiology of the brain.” 

30 Interesting and discriminating accounts are those of E. G. Boring, A History 
of Experimental Psychology , 1929, chap, iii; M. Bentley, “The Psychological Ante- 
cedents of Phrenology,” Psychol. Monog. y 1916, No. 92; C. Blonde!, La psychophysi- 
ologie de Gall , 1913; McQ. De Grange, The Science of Individuality , 1923; H. D. 
Spoerl, “Faculties versus Traits: the Solution of Franz Joseph Gall” Char. & Pers. t 
1936, 4, 216-231. 

31 These assumptions, or “laws” as they were often called, are stated briefly in 
the following way: 

(a) mental faculties are innate; 

(b) the brain is the organ of mind; 

(c) the form and size of the brain are distinguishable by the form and size of 
the head or skull; 



though we can afford to disregard the many spurious features of the 
system along with its curious history, there still remain three positive 
contributions in Gall’s work deserving respectful consideration today. 

In the first place, there is an acceptable postulate underlying Gall’s 
organology, namely, that mind and body are not two independent 
entities, but are inextricably related. In one way or another virtually 
all of the contemporary work on personality shares with Gall this 
implicit faith in psychophysical parallelism. Furthermore, it was Gall 
himself who brought this philosophical conviction into fashion among 
characterologists. To be sure, it does not specifically advance our 
knowledge to hold that an individual’s psychic attributes are rooted 
in some way in the physiological functions of his body. The view, 
however, becomes important when it is regarded as a framework for 
more detailed theorizing and research. Probably no modern investi- 
gator doubts that through scientific discovery patterns of personality 
will be found to parallel patterns of somatic response. “Engrams,” 
“neurograms,” “physiological vectors,” “visceral-cortical tracts” are 
some of the terms nowadays proposed to aid in establishing this cor- 
relation. It was Gall’s intention to create through his biological studies 
a conceptual unification of the mind-body relationship as an aid in 
understanding human personality. Most investigators of the present 
day would like to do the same thing. 

In the second place, phrenology rendered timely service by calling 
attention to the phenomenon of individual differences. Psychology 
in the early nineteenth century was interested almost exclusively in 
mind-in-general. Gall recognized that human beings differ widely in 
their personal tastes and mental qualities, and yet he hoped, as many 
psychologists of the present day hope, to discover basic variables that 
might account for the apparently limitless varieties of human per- 
sonality. He struggled persistently to establish a final list of the essen- 
tial, or as he called them the “primitive,” characteristics of human 

(d) the mind possesses distinct faculties and the brain is composed of distinct 
organs, and each mental faculty is manifested through a distinct cerebral 

(e) the size of each organ can be estimated during life, and the size, other 
things being equal, is a measure of power; 

(f) each organ, when predominantly active, impresses the body with certain 
uniform attitudes and movements, called its “natural language.” (From 
George Combe, Lectures on Phrenology, 1847, P* 63.) 

Of these assumptions only (b) arid (f) are by any stretch of the imagination 
acceptable to modem psychologists, and these not in the sense Gall intended. 


What is more, in seeking these radical elements (which turned 
out to be the famous twenty-seven faculties) Gall employed an em- 
pirical method. He was, as he thought, decidedly positivistic in his 
procedure. He compared innumerable skulls, studied genius, labored 
in asylums for the insane, reasoned inductively, and with the most 
honest of intentions criticized and revised his own work constantly. 
He obeyed, much more faithfully than most of his contemporaries, 
the canons of scientific method, and yet he committed absurdities. 
His failure should be a warning to the devotees of positivism that a 
method in itself is never a certain guarantee of truth. 

Today with different instruments a similar positivistic search is 
being made for the “primitive” or basic elements of personality. 
Instead of being the audacious quest of a single investigator, handi- 
capped by an assistant as headstrong as Spurzheim, the modern proj- 
ect has been the subject of a call for international co-operation of 
all the psychologists who are interested in applying statistical methods 
to the problem. The method, like Gall’s, is to be empirical, but every 
assistance of modern tests and experimentation together with the aid 
of mathematics are to be utilized. 32 It is “factors” rather than “facul- 
ties” that are sought, but the intent of the search is similar. Whether 
this modem endeavor will meet with greater success than Gall’s re- 
mains to be seen. Part III of this volume will define some of the 
serious problems that must be faced in any attempt to discover the 
ultimate unitary traits of human nature. 

The third meritorious contribution of Gall’s work is the most im- 
portant, but least understood of all. It has to do with the nature of the 
units of personality (the “faculties”) with which he dealt. In contem- 
porary psychology there is, of course, a prejudice against faculty 
psychology in all its forms— an unfortunate fact since such a totalized 
prejudice prevents balanced criticism and discriminating judgment. 
As a matter of fact there are many varieties of faculty psychology, 
and Gall’s is seriously misrepresented when it is identified with other 
faculty psychologies current in the early nineteenth century. Cast by 
his critics into the same camp as Wolff, Stewart, Reid, and Plutche- 
son, Gall has been unfairly sentenced with them to oblivion. 33 

82 C. Spearman, X th Congress of Psychology, Copenhagen, 1932. A recent re- 
port on this co-operative research is K. J. Holzinger, “Recent Research on Unitary 
Mental Traits” Char . & Pers., 1936, 4, 335-343. 

33 Wundt, for example, claimed that Gall produced simply a physiological cari- 
cature of Wolff’s faculties (Qehirn und Seele, 2d ed., 1906, 145-148). Similarly, James 
asserted that Gall “took the faculty-psychology as his ultimatum on the mental side, 



The charge that Gall was imitating the faculty psychologists of 
his time is false. Time and again he inveighed against Wolff, and 
other continental psychologists. He regarded their a priori lists of 
faculties as totally worthless, and pursued what he thought was a 
strictly empirical, and far sounder, method of discovering the really 
radical units of character. For one thing, he sought faculties that 
would be independent of one another. No other faculty-psychologist 
claimed independence for his faculties, and none sought as Gall did 
to establish the faculties inductively and empirically through a direct 
examination of innumerable individual cases. But the chief difference 
between Gall’s psychology and that of the others lay in the fact that 
he sought with the aid of his “faculties” to account for the differences 
between men. “We need faculties,” he wrote, “the different distribu- 
tion of which shall determine the different species of animals, and the 
different proportions of which explain the difference in individuals.” 34 
All other faculty psychologists aimed to establish universal faculties, 
such as would account for the mental operations of all men. They 
were not interested in individual differences at all, certainly not in 
the problem of the organization of unique personalities. In envisaging 
this problem, Gall was a century ahead of his time. 

The point at issue is important enough to deserve an illustration. 
“Perception,” for example, does not appear in Gall’s list of faculties; 
it does appear in Reid’s and Stewart’s (cf. lists in table p. 84). It 
seemed self-evident to the latter that perception is a basic power of 
all minds, and so it is. But for Gall, the very universality of Percep- 
tion disqualified it as a radical (differentiating) faculty. He would 
say that such a basic process is common to the exercise of all the 

and he made no farther psychological analysis” (Principles of Psychology , I, 27). 
How could these famous founders of modern psychology reconcile their charges 
with Gall's own repeated condemnation of faculty psychology as it was known to 
him? (CL Gall, On the Functions of the Brain, Vol. III, pp. 82-86.) 

James lays the disrepute of phrenology largely to the vagueness and vastness 
of its faculties which, he thought, not only lacked analytic finesse but were con- 
ceived as so many separate self -active “souls” or homunculi. To James, as to the 
modem specificists, the faculties should have been analyzed into smaller, sensori- 
motor, elements. “A science of mind must reduce such complex manifestations as 
‘philoprogenitiveness' to their elements” (Principles, % 27). 

It is true that Gall did not analyze his 27 functions into elements, but his fail- 
ure to do so is the very reason why his list of human qualities seems more in keep- 
ing with the structured character of personality as we know it in actual life than 
do the motley assortments of habits, conditioned reflexes and other sensorimotor 
elements out of which recent psychology has vainly attempted to create a scientific 

u Gall, On the Functions of the Brain, Vol. I, p. 88. 



faculties; its role is merely attributive; it is not a characterological 
unit. On this ground he repudiates not only Perception, but many 
other universal faculties proposed by his contemporaries. He wrote, 
for example, “W e nowhere find that a man or a woman has become 
celebrated by the Understanding and the Will, by Attention, Com- 
parison, Desire. . . . Every man, except an idiot, enjoys all these 
faculties. Yet all men have not the same intellectual or moral charac- 
ter.” 35 What is needed, he concludes, is “primitive” units which will 
account for the distinctions, and not merely the resemblances, be- 
tween human beings . 36 

Most of Gall’s disciples did not perceive the significance of his 
distinction between the nomothetic, or universal attributes of mind, 
and the concrete, differentiating, faculties which he so arduously 
sought to discover. His followers, of whom Spurzheim is a good 
example, were attracted by the pseudo-practical applications of 
phrenology. An exception to this rule is von Struve who re-states 
the issue clearly: “When I accurately specify each of the thirty-five 
phrenological qualities of a man , 37 I have laid the foundation for a 
graphic account of character from which its direction can be traced 
both in its larger and in its detailed aspects, and in relation to intellec- 
tual and moral factors. But when, with the old school, I speak in 
general only of the life of experience, feeling, and the like, they 
throw little light upon the matter.” 38 Another shrewd reader of Gall 
was Thomas Hyde, who, as a senior in Harvard College, presented 
an honor’s thesis to William James comparing the merits of phre- 
nology and psychology, then a young experimental science. He con- 
cluded that, “the establishment of primitive powers was chiefly the 
work of phrenology,” whereas psychology “seems better able to de- 
tect the general or universal rather than the specific or individual.” 
Hence, he concludes, “after a careful consideration of the claims of 
each, we gave our adherence to phrenology.” 39 This essay, we are 

35 Loc. cit, 

36 The opposition between “universal faculties” and “primitive faculties” is diS' 
cussed more fully by H. D. Spoerl, “Faculties or Traits: The Solution of Franz 
Joseph Gall,” Char. & Pers 1936, 4, 216-231. This study shows an excellent appre- 
ciation of the significance of Gall’s work for the psychology of personality. The 
readiness of most psychologists to confuse Gall’s conception of the faculties with 
those of his predecessors is due to their pre-occupation with mind-in-general. They 
fail to understand that Gall’s ultimate interest was mind-in-particular. 

37 Gall’s original list of twenty-seven faculties was increased to thirty-five by 
Spurzheim and to forty-two by Combe and other followers. 

38 G. von Struve, Phrenologie in mid aasserhalh Deutschland, 1843, p. 51. 

39 Thomas A. Hyde, How to Study Character, 1884, pp. 170 f. 



told, was favorably commented upon by James, but the crucial dis- 
tinction it drew between universal and characterial faculties was ap- 
parently unappreciated by James, for it did not find its way into his 
own critique of phrenology. 

Thomas Reid, 17S0 

Active Powers 
Maintenance of Habits 

Hunger and Thirst 

Instinct of Imitation 


Desire for Power 

Desire., of Knowledge 
Conciliate Affection 
■ Gratitude 

Pity and Compassion 

Esteem of the Wise and Good 

Sexual Affection 
Public Spirit 

Rational Resentment \ 

Animal Resentment ) 

Transcendent Good 


Imagination ( — invention) 


Intellectual Powers 
The Five Senses and Their 
Faculty of Perception 

Size and Novelty 


Judgment and Reason 



Dugald Stewart, 

Franz Joseph Gall, 1810 

Moral Taste 

Active Powers 

Propensity to Action and 

Hunger and Thirst 

Acquired^ Appetite .for Drugs 
Desire of Society. 

Instinct of . Imitation 

Self-Love ) 
Self-Confidence j 
Desire of Knowledge 
Parental Affection 
Filial Affection 


Universal Benevolence 

Desire of Esteem 



Sexual /Affection 


Desire of Superiority 



Sense of Duty 



Decency, Regard to Character 

Instinct for Construction 
Sense of Similarity and \ 
Contrast j- 

Sense of the Ridiculous J 
Memory for Colors 

Intellectual Powers 
The Five Senses and Their 
Faculty of Perception 

Judgment and Reasoning 




Moral Taste 

Association of Ideas 

Determinate Faculties 

Instinct of Generation (i) 

■Mimicry, Imitation (25) 
Verbal Memory (14) 
Vanity, ..Ambition (g) 

Pride, Self-Esteem. (8) 

Educability (11) 

Love of Offspring (2) 

Good Nature (.24.) 
Friendship, Attachment {3) 

Courage, Self-Defense (4) 

Theosophy, Religion (26) 

Firmness of Character (27) 
Poetry (23) 

Mechanical Aptitude (39) 
Wit (22) 

Sense of Colors (16) 

Music (17) 

Wish to Destroy (5) 
Cunning (6) 

Sentiment of Property (7) 
Cautiousness (10) 
Mathematics (18) 

Memory for Persons (13) 

Local Memory (12) 

Memory for Languages (15) 

Comparative Sagacity (20) 
Metaphysical Depth (21) 


is irJparentheses'; lH&FTOed’n?mber?)” VerSity P ‘ GalI ’ S numt,er!n S of tIle faculties 



As clear as Gall’s intention was, it must be admitted that he was 
not uniformly successful in his achievement. Some of the faculties 
listed in Spoerl’s table all too obviously parallel the Scottish faculties, 
and thereby reflect the same abstractions from the generalized mind, 
rather than clear-cut characterial units. Memory for languages is one 
example; color sense is another. Gall would undoubtedly reply that 
he had accepted such of these faculties as seemed to him to be quali- 
ties whose quantitative variation did actually account for differences 
in character; when intense, these attributive faculties actually operate 
as characterial (differentiating) faculties as well. Whatever the merits 
of this reply, Gall is guilty of confusing passive (intellectual) facul- 
ties, viz., aptitudes and skills, on the one hand, with active, conative 
integrations of character, on the other. Actually it is only the latter 
that fit his requirements for “primitive” units. 

It is easy enough in retrospect to see what Gall should have done 
in order to accomplish his admirable purpose of constructing a psy- 
chology of individuality. He should have rejected the term “faculty” 
as too heavily freighted with a meaning wholly foreign to his own. 
He should have then distinguished not only in theory between the 
universal and the characterial method of analysis, but should have 
followed his conviction more consistently in practice. Only by so 
doing could he show that the former faculties represented an undesir- 
able abstraction, and that the latter followed in principle the concrete 
personal organizations of nature. He would have had to eliminate in- 
tellectual aptitudes and talents from his list more completely than he 
did, and confine himself to such personal-conative systems as are 
nowadays referred to as interests, sentiments, values, and traits. This 
step would have led him away from his two other most serious errors, 
namely, the assumption that each of the units he chose is independent 
of all other units, and the equally false assumption that every radical 
quality of character is inborn and “resists education.” 

Gall’s belief that personality is naturally organized into more or 
less systematized dispositions, each of which expresses individuality of 
adaptive behavior, is altogether acceptable. If shorn of its many false 
embellishments (the myth of corresponding organs, its nativistic bias, 
its numerical listings, and its occasional and inconsistent concessions 
to the prevailing faculty psychology of the day) it would then 
serve as an admirable starting point for a modern psychology of per- 
sonality founded upon a clear conception of the nature of traits. 




Under the title of “Ethology” John Stuart Mill proposed the 
formation of “an exact science of human nature .” 40 It should deal, he 
maintained, with human character, and be established upon the secure 
foundations of a general and abstract science of psychology. Its ma- 
terial should consist of the empirical wisdom of common sense, de- 
scriptions of personal patterns of conduct, but its explanatory prin- 
ciples must be derived from the science of psychology. This distinc- 
tion between descriptions of character and explanations of character 
is of considerable importance . 41 

Only the explanatory principles in characterology, according to 
Mill, can be derived from the science of psychology. Its data must be 
drawn from life and not from experiment. Experimentation with hu- 
man character, he thinks, is impossible. To prosecute such studies it 
would be necessary to bring up and to educate individuals in com- 
plete isolation, with every conditioning factor known and controlled 
from infancy to a mature age. (Mill’s requirements for an experiment 
were much more exacting than those of the present day!) There are, 
however, innumerable collections of proverbial wisdom concerning 
human characteristics which are to be found in maxims, adages, and in 
the literature of all ages. These empirical generalizations affirm tend- 
encies, not facts. It is, for example, said that bodily strength tends to 
make men courageous; not that it always makes them so; it will do 
so only if no counter-influence intervenes. Now an accumulation of 
such empirical wisdom becomes valuable when it can be referred to 
psychological laws, and in this way be checked by placing it in a 
framework of “causal” explanations. “Unless we have resolved the 
empirical law into the laws of the causes on which it depends, and 
ascertained that those causes extend to the case which we have in 
view, there can be no reliance placed in our inferences.” 42 The really 
scientific truths, then, are not these empirical laws, but the causal laws 
which explain them. 

Mill realizes that general psychology alone is not equipped to deal 

40 J. S. Mill, System of Logic, 1843, Book VI, chap. v. 

41 The distinction is the same as Lewin draws between the phenotype (“the im- 
mediate perceptible appearance”) and the genotype (“genetic conditioning”); cf. 
p. 16. Throughout the psychology of personality die opposition between appearance 
and cause constandy recurs. 

42 Bk. VI, chap, v, sec. 2. 


with character. It merely supplies a statement of laws useful in ex- 
plaining the development of character. This view is the same as 
Wundt’s (pp. 6 f.), who regarded the science of “characterology” as 
a supplement to general psychology. Mill wrote before Bahnsen had 
coined the term “characterology,” and therefore was unable to profit 
from it. He says, 

A science is thus formed, to which I would propose to give the 
name of Ethology, or the Science of Character, from f|frog, a word 
more nearly corresponding to the term “character,” as I here use it, 
than any other word in the same language. The name is perhaps 
etymologically applicable to the entire science of our mental and 
moral nature; but if, as is usual and convenient, we employ the name 
Psychology for the science of the elementary laws of mind, Ethology 
will serve for the ulterior science which determines the kind of char- 
acter produced in conformity to these general laws, by any set of 
circumstances, physical and moral. 43 

Referring to the accumulated maxims concerning character, and 
to the great advances of general psychology, Mill asserts that the 
creation of this new science has at length become practicable. “The 
empirical law's, destined to verify its deductions, have been formed in 
abundance by every successive age of humanity, and the premises for 
the deductions are now sufficiently complete.” 44 

What were these premises which were sufficiently complete? 
What causal principles did psychology at the time of Mill have to 
offer? When this question is asked it becomes clear immediately why 
Ethology made no advance for fifty years after Mill published his 
program. Associationism, the principle by which fragmentary states of 
consciousness aroused other fragmentary states, was the sole “explana- 
tory” tool of psychology, and woefully inadequate to account for the 
galaxy of human interests, motives, conflicts, and passions which are 
the essential forces in the formation of character. Psychology in Mill’s 
time was intellectualistic, Apollonian, and not until the influence of 
Schopenhauer, Darwin, Freud, and McDougall had altered its point 
of view radically, training its vision upon the irrational motives of 
men, were the premises sufficiently complete to permit a realization 
of Mill’s proposal. 

It was Alexander Shand who put Mill’s method to its only ade- 

43 Ibid., sec. 4 . 

44 Ibid., sec. 6. 



quate test. Following Mill’s instructions he assembled countless prov- 
erbs, maxims, and literary statements concerning human nature. But 
for his “premises for deduction” he accepted not the laws of associa- 
tion but the theory of sentiments. 48 Apart from his use of this new 
type of psychological explanation, his method follows Mill’s proposal 
exactly. He deduces 144 laws for the “foundations of character,” un- 
der the guidance of his basic law that “Every sentiment tends to 
form a type of character of its own.” 4S The individual laws refer to 
the typical courses which the major sentiments take in human life, for 

(Law 1 16) Hope tends always to make the future appear 
better than the present. 

He then “verifies” each law by common wisdom and adages to con- 
form to Mill’s requirement that “verification a posteriori must pro- 
ceed pari passu with deduction a priori .” 47 In the case of this par- 
ticular law, Shand offers among other proofs the authority of 

True hope is swift, and flies with swallow’s wings, 

Kings it makes Gods, and meaner creatures kings. 

The outstanding weakness of Shand’s procedure is his arbitrary 
selection of supporting proverbs and maxims. However much of an 
advance it was to insist upon dynamic and emotional factors as the 
foundations of character, reliance solely upon sentiments limited un- 
duly the genotypical background with which psychologists of char- 
acter must work. As for proverbial wisdom, it is notoriously am- 
biguous. Every proverb affirming one type of occurrence, seems to 
engender another negating it. It is said that “as a twig is bent, so 
inclines the tree,” but also that “a young monk makes an old devil.” 
Goethe wrote, “The weak often have revolutionary sentiments”; but 
G. B. Shaw has maintained that “a man who is not a revolutionary by 
twenty is an inferior.” 

The positive contribution of Shand is his recognition of systema- 

45 In Mind, in 1896 Shand published an article giving the outlines of his theory, 
but it was not until 191; that his book Foundations of Character appeared. Between 
these dates Stout and McDougail had made important contributions to the theory of 
sentiments. Cf. W. McDougail, “Organization of the Affective Life. A Critical Sur- 
vey,” Acta Psychologica, 1937, 2, 233-346. 

46 Op. cit., p. 123. 


tized emotional dispositions as the functional units of which the per- 
sonality (or as he prefers to call it, the character) is composed. Lik e 
Gall, he sought units that would differentiate men from one another 
in respect to those complex affective and conative functions which 
are the ultimate systems of adaptive behavior. His interest basically, 
however, is nomothetic, for he hopes through the compounding of 
the same common sentiments in all men, with the aid of the same one 
hundred forty-four laws, to account for all manifestations of individ- 
uality. His approach is significant for his selection of the dynamic 
unit of the sentiment, but it is not completely personalistic; and his 
neglect of all psychological aids other than the concept of sentiment 
is short-sighted. 

Similar to Shand’s, but resting upon a more adequate psychologi- 
cal groundwork, is McDougall’s theory of personality. Elere again 
“sentiment” is the cardinal concept. The sentiment is itself a com- 
plexly organized affective tendency, springing from instinct and emo- 
tion, but attached by experience to a certain object or class of ob- 
jects. In themselves the various sentiments may obstruct one another 
or conflict with one another unless they are brought into one single 
system within which their impulses are harmonized. This embracing 
integration is the “character,” and is achieved through the develop- 
ment of a master sentiment of “self-regard,” which takes the form of 
a self-conscious devotion to certain selected ideals with which the 
person identifies himself. Not only does this devotion to certain ideals 
dominate and harmonize all other sentiments, but it is capable of ex- 
tension and change so that a continuous and consistent growth of per- 
sonality is assured. Failure to organize the system of sentiment-units 
into a hierarchy with the sentiment of self-regard “at the top,” leads 
to mental conflict, neurosis, and in extreme cases to psychosis . 48 

Two other writers of the past century, preceding Shand and 
McDougall, placed stress upon the role of dominant interests or “rul- 
ing passions” in conferring form and unity to personality. One of 
these, Alexander Bain, was well acquainted with the history of char- 
acterology. He wrote both appreciatively and critically of his prede- 
cessors, and even prepared an original translation of the Characters 
of Theophrastus . 49 Mill’s provocative plan for Ethology inspired him 
to offer an alternative program for characterology which would be, 

* 8 W. McDougall, Outline of Abnormal Psychology, 1926, esp. pp. j2jf. 

49 Alexander Bain, On the Study of Character? 1862, 


he thought, “more in accordance with the present state of our knowl- 
edge of the human constitution.” 

The foundations of Bain’s doctrine were the three venerable fac- 
ulties of Emotion, Volition, and Intellect, derived originally from 
Plato’s tri-partite division of the human soul. By Bain these faculties 
were viewed as so many channels for the flow of “psychic energy.” 
Each man has a characteristic amount of this energy, which may 
either be wasted or turned to account. When it is not merely ex- 
pended at random (as in play) it is purposefully directed into emo- 
tional, volitional (muscular) or intellectual activity. A man’s nature 
thus is determined by the predominance of one or the other of these 
three channels. “Human nature being limited, if one’s vitality runs 
very much to the active organs, less will go to the other parts.” 50 
This “steam-boiler” conception results inevitably in a doctrine of 
types, with the result that there are men who are mental , others who 
are motor (“men of action”), and still others who are vital (sensual). 

Popular and commercial characterology of today employs this 
same three-fold distinction, often supplementing the types with reck- 
less physiognomic parallels. For instance, it is said that the mental 
type possesses an elongated or triangular face, the motor type, square 
and aggressive features, and the vital type, a round and flabby coun- 

Bain’s fallacy lay in assuming a constant amount of “psychic 
energy” which if drained into one channel would necessarily flow in 
deficient qualities through other channels, shaping the personality 
in accordance with its flow. The resulting typology seems too scanty 
in its permutations and unconvincing in its selection of channels and 
in its theory of energy. It has, however, the merit of recognizing as 
the central problem in the psychology of personality the study of 
dominant motives and interests. 

The other characterologist who emphasized the importance of 
the dominant interest or “ruling passion” is Charles Fourier (1772- 
1837). His analysis of human passions served him as a basis for the 
elaborate social philosophy and doctrine of reform for which he is 
famous. 51 

The intricacies and eccentricities of Fourier’s classification of 
human types are fascinating. He was obsessed by the possibility of 

50 Alexander Bain, op. cit., p. 201. 

51 Charles Fourier, The Passions of the Human Soul, trans. i8ji, two volumes 
See also A. A. Roback, The Psychology of Character, chap. x. 


distributing mankind into regional and vocational groups according 
to his own elaborate scheme of character types. The fundamental 
“passions” of men are of three classes (sensuous, affective, distribu- 

Figure io 

Physiognomic Parallels to Bain’s Types (as offered by popular, 
“character analysis”). The oversimplification is apparent from the 
selection of extreme physiques, and from the use of accentuating 
boundary fines and goatee. No proof is offered for the validity of the 

Reprinted by permission from Psychology , 1925, 4, p. 41. 

tive), but these must further be divided into 12 orders, 32 genera, 
1 34 species, 404 varieties. When recast into types of character there 
are 810. Two-thirds of these types are “monogynes,” that is, indi- 
viduals who have only one dominant motive, a few are “digynes” 
(with two dominant motives). Rarer and rarer become men of mul- 
tiple motives and the attendant insight and wisdom, though occasion- 


ally (twice in 29,222 cases) an “omnigyne is encountered. By fur- 
ther computation, a condition of character exists four times in three 
billions of people which must be called the Super-omnigyne. Speak- 
ing modestly of his own accomplishment in discovering this classifica- 
tion, Fourier writes, “This exalted degree has this singular property 
of discovering, almost by inspiration, the laws of harmony. I must 
necessarily be of this degree, since I have arrived at it without any 

Men may be classified not only in respect to the number but also 
according to the nature of the leading motives they possess. Motives 
are either “dominant” or “tonic.” The former are the guiding pur- 
poses, and the latter confer color, flavor, and style on the execution 
of the purpose. A certain king of Saxony, Fourier tells us, used to 
write on the last day of the year a plan for his activities during each 
day of the ensuing year. “On the first of March I shall go to the 
hunt; on the second of April, I shall spend the day fishing; on the 
third of May I shall hold a Council of State. . . .” According to 
Fourier this methodical king had a “dominant” of ambition, and a 
tonic of “stability.” This quaint doctrine contains one lesson of im- 
portance. No psychology of personality can be written exclusively 
in terms of purposes, goals, needs, or instincts. These “dominants” 
leave completely out of account the manner or personal style through 
which the purpose is achieved. There are many ways in which men 
may express ambition. Personality involves not only purpose, but also 
certain habitual and individual modes of attaining the chosen goals; 
not only the dominant determinants, but the tonic determinants as 
well. The necessity for such dual determination was clearly seen by 
Fourier; it is recognized likewise in Stem’s specification of both the 
i Richtungsdispositionen and the Riistungsdispositionen in personality. 
The relation between such “driving traits” and “directive traits” is 
discussed more fully in Chapter XII. 

Fourier thought that he had invented a “science of concrete 
charm, applicable to the wants of the whole human race,” that would 
“put an end to its long endured miseries.” If society were organized 
so as to employ the talents of its members according to his plan of 
classification, great increases in production would result. For example, 
the proceeds of six months’ sale of hen’s eggs, when poultry raising 
was in the hands of the proper monogynes, would pay off the British 
national debt! 

Eccentric as it is, Fourier’s scheme has three merits. In the first 



place, it serves as a warning against over-elaborate typologies. Divi- 
sions and sub-divisions soon get out of hand, and neither the author 
nor the reader can work with a typology whose ramifications exceed 
his own range of manipulation. The temptation with every typology 
as soon as its inadequacies are apparent is to elaborate it through sub- 
division. The process ultimately leads to the swamps of confusion. 
A second, more positive, contribution is Fourier’s emphasis, in com- 
mon with Shand, McDougall, and Bain, upon the dominant motive as 
the integrative and distinguishing factor in personality. 52 Lastly, 
Fourier’s distinction between dominant and tonic “passions” recog- 
nizes both as important phenomena, not wholly independent of one 
another, yet by no means identical. 


During the nineteenth century the literature of characterology 
was rich in observation and hypothesis, but no writer prior to 1884 
proposed to apply the newly developed experimental methods of psy- 
chology to the study of personality. Physiognomists and phrenolo- 
gists, it is true, had spent much time examining people, but for their 
purposes it mattered little whether the people they studied were dead 
or alive; in no case did they conduct their examinations under what 
today would be called “controlled conditions.” 

John Stuart Mill, it will be remembered, repudiated the experi- 
mental method as unsuited to the study of personality. His science 
of Ethology could have been pursued by a recluse without so much 
as a glance at another human being, nor even at himself, provided he 
had access to the world’s literature of proverbs and to treatises in 
theoretical psychology. Even the rapidly growing interest in experi- 
mental psychology, which in 1879 materialized in a psychological 
laboratory at Leipzig, was concerned not at all with personality. The 
nearest approach it made was the “personal equation” in reaction 
times, and in other equally simple, nomothetic functions. 

It was Sir Francis Galton (1822-1911), the pioneer in so many 
psychometric fields, who for the first time proposed explicitly that 
the standards of experimentation be applied directly to the study 

52 The aphorism, “A man is what he loves/" fits this type of theory very well. 
It is endorsed by many characterologists today. (Cf. Havelock Ellis, The Dance of 
Life, 1923, p. 336; also E, Stern, “New Ways of Investigating the Problems of Per- 
reality,” Psyche, 1923, 3, 358-366.) . 


of personal, particular forms of behavior. After satisfying himself 
through his previous work on genius and twins, and through his in- 
trospective studies of his own behavior that all of man’s actions “seem 
clearly to lie within the province of normal cause and consequence,” 
he reached the conclusion that “the character which shapes our con- 
duct is a definite and durable ‘something,’ and therefore that it is 
reasonable to attempt to measure it.” 5S One method of measurement, 
advocated by him, was the rating of complex human qualities. School- 
masters especially, he pointed out, had an enviable opportunity to 
count the frequency and estimate the intensity of their pupil’s re- 
sponses of anger, fear, loyalty, and ambition. Norms were to be estab- 
lished for the development of character at successive ages. Many 
statistical methods, especially those of correlation, likewise made 
available for the first time by Galton, were to be employed in these 
investigations. This advance in empiricism, now accepted by the great 
majority of investigators, alone would have entitled Galton to a pre- 
eminent position in the history of characterology. But his proposal to 
employ actual experiment is still more radical. 

Observation is a slow process, especially while waiting for critical 
episodes of life to appear. But “emergencies need not be waited for; 
they can be extemporized; traps, as it were, can be laid.” Galton gives 
the following example, 

Thus when two persons have an “inclination” to one another, they 
visibly incline or slope together when sitting side by side, as at a 
dinner-table, and they then throw the stress of their weight on the 
near legs of their chairs. It does not require much ingenuity to arrange 
a pressure gauge with an index and dial to indicate changes in stress, 
but it is difficult to devise an arrangement that shall fulfill the three- 
fold condition of being effective, not attracting notice, and being 
applicable to ordinary furniture. I made some rude experiments, but 
being busy with other matters, have not carried them on, as I had 
hoped . 54 

This humorous, but nonetheless reasonable, proposal marks the 
beginning of experimental characterology. The special problem with 
which it was concerned, expressive movement, has, regrettably 
enough, advanced little since Galton’s ingenious suggestion. But his 
faith in experimental procedures has, in the main, triumphed. 

53 F. Galton, “Measurement of Character,” Fortnightly Rev., 1884, 42, 179-185 

54 Op, cit. } 184. 



The invention and spread of experimental techniques after Gal- 
ton’s time are scarcely yet matters of history. Experimentation is the 
very tissue of contemporary research, and its story must be told in 
every subsequent chapter of this volume. 


Out of the long and varied past this chapter has selected only 
those characterological doctrines that have some special significance 
for the contemporary study of personality. In a few cases ancient and 
persistent blunders have been discussed, but for the most part posi- 
tive contributions have been recorded. 

Theophrastus saw that the small every-day acts of men are re- 
markably congruent with one another. To him the “elements” of 
personality were not specific and fragmentary; on the contrary, the 
only unit of any consequence to him was the dominant trait from 
which a man’s every habit derives its significance. La Bruyere failed 
to specify, as Theophrastus did, what dominant traits his characters 
possessed, but he drew portraits reflecting an equally consistent style 
of life. The physical qualities, the manners, bearing, and address of 
his subjects constitute an intricate but unified pattern. 

With greater freedom, poetry, drama, and fiction portray not 
only the consistencies but the inconsistencies of human nature. So art- 
ful are its masterpieces of characterization that psychology, by con- 
trast, seems clumsy and inept. But standards of objectivity and verifi- 
ability strengthen the case for psychology. Clumsy exactness is a 
wholesome antidote to undisciplined felicity. Either alone is one- 
sided; but taken together, the two methods provide a well-rounded 
equipment for the student of personality. 

Humoral psychology, the longest standing of all doctrines of 
human nature, owes its longevity partly to the uncontroverted claim 
that there is a correspondence between the chemistry of the body 
and emotional make-up, and partly to its implicit recognition of cer- 
tain fundamental dimensions of temperament. 

Physiognomy too maintains that some relationship between tem- 
perament and the skeletal structure should be expected. Today we 
would say that this correspondence is due to the dependence of both 
physical development and emotional excitability upon the action of 
the hormones. In recent times Kretschmer and his many followers 
have proposed a nativistic physiognomy based upon this logic; but 


they overlook the influence of experience upon muscular develop- 
ment. Beside the constitutional type we must know the significance 
for physiognomy of the common instinctive patterns of emotional 
expression, the influence of custom and convention, and the effect of 
personal experience in determining posture, and also movement. 
Lavater held that ultimately all expressions must be consistent with 
one another because all alike proceed from the unity and harmony 
of a man’s own inner nature. 

Phrenology stresses even more specifically the correlation between 
a man’s personal qualities and his bodily structure. The correlations 
are most certainly not what Gall believed them to be. Yet Gall’s car- 
dinal error, “organology,” has blinded critics to his positive contribu- 
tions, leading them to read in his work the prevalent faculty psy- 
chology of his time, which he had, in fact, expressly repudiated. Gall 
must be credited not only with helping to establish the hypothesis of 
psychophysical parallelism from which most modern work on per- 
sonality proceeds, and with creating a new interest in the psychical 
differentiations between men, and with the adoption of positivistic 
methods (however far astray they led him), but he must be credited 
especially with the important distinction between universal (nomo- 
thetic) faculties and characterial (primitive) faculties. Seeking a psy- 
chology to account adequately not for the similarities among men but 
for the differences between them, he arrived, somewhat confusedly, 
to be sure, at the first systematic conception of traits. 

In proposing Ethology as an exact science of human nature, Mill 
envisaged clearly the interdependence of psychology, literature and 
common sense. Like Wundt, Mill would base his characterology 
firmly upon psychology, but allow it new methods and greater lati- 
tude. Shand and McDougall find in the “sentiments” an adequate 
basis for a psychology of personality, a more satisfactory foundation 
than the older associationism. Bain offered a still simpler conception 
of the energic flow, later revived and modified by Jung and by 
Kempf. Gradually conation and emotion came to bear the central 
emphasis. Now with the added support of Freud it is unlikely that 
the study of personality will ever slip back into the intellectualistic 
formulations of the past. 

Fourier’s eccentric doctrine acts as warning against elaborate 
typologies. Fourier, however, does call attention to the rich variety 
of motives playing decisive roles in the development of personality. 
Furthermore, the distinction he draws between “dominant” and 



“tonic” traits anticipates an important modern distinction between 
the driving motives of life and the directive or expressive styles of 

Fifty years ago an entirely new era began in characterology when 
Sir Francis Galton argued for the first time for direct experimenta- 
tion, at the same time contributing many novel and now indispensable 
techniques. Mill had declared experimentation in the field of person- 
ality to be impossible, and many skeptics have since agreed with him. 
Nevertheless, it is not Mill’s view but Gal ton’s that has prevailed, and 
that seems destined to dominate the psychology of personality during 
the twentieth century. 

Part II 




Nature, then, has generated and fashioned man’s body in 
such a way that some parts of it were perfect at birth, others 
were formed as its age increased, without much use of external 
and adventitious aids. Now in other respects she made the 
mind as perfect as the body, endowing it with sense capable of 
perceiving things, so that little or no assistance of any sort was 
needful to supplement them. But that faculty which is highest 
and most excellent in man she left lacking ... she furnished 
merely the rudiments; nothing more. 


Only the rudiments of that “which is highest and most excellent 
in man” are given at birth. The fully fashioned social and moral being, 
the developed adult personality, waits upon the process of growth. 
The nature of growth is the critical problem for the psychology of 
personality. For above all else it must know how the biological or- 
ganism it finds at birth becomes transformed into the adult person 
able to take his place in the highly complex social activities of the 
civilized world surrounding him. 

As yet no pooling of the resources of physiological, genetic, com- 
parative, and social psychology can give a complete answer. There 
are gaps as well as contradictions in the evidence. Knowledge of 
heredity is meager; the problems of instinct, maturation, and learning, 
though fairly well formulated, are unsolved. Adding to the difficulty, 
what seems to hold true for one child does not always hold true for 
another— the weights of the factors involved in growth apparently 
vary from case to case. 

Take, for example, a typical problem of the influence of the 
home environment. Great though it undoubtedly is, it seems to op- 
erate in antithetical ways. The novelist, Sinclair Lewis, in the fol- 
lowing passage about two brothers has brought out this issue most 



“My father,” said Ora, “was a sloppy, lazy, booze-hoisting old 
bum, and my mother didn’t know much besides cooking, and she was 
too busy to give me much attention, and the kids I knew were a 
bunch of foul-mouthed loafers that used to hang around the hoboes 
up near the water tank, and I never had a chance to get any formal 
schooling, and I got thrown on my own as just a brat. So naturally 
I’ve become a sort of vagabond that can’t be bored by thinking about 
his ‘debts’ to a lot of little shop-keeping lice, and I suppose I’ni in- 
clined to be lazy, and not too scrupulous about the dames and the 
liquor. But my early rearing did have one swell result. Brought up so 
unconventionally, I’ll always be an Anti-Puritan. I’ll never deny the 
joys of the flesh and the sanctity of Beauty.” 

“And my father,” said Myron, “was pretty easy-going and always 
did like drinking and swapping stories with the Boys, and my mother 
was hard-driven taking care of us, and I heard a lot of filth from the 
hoboes up near the water tank. Maybe just sort of as a reaction I’ve 
become almost too much of a crank about paying debts, and fussing 
over my work, and being scared of liquor and women. But my rearing 
did have one swell result. Just by way of contrast, it made me a good, 
sound, old-fashioned New England Puritan.” 1 

The same fire that melts the butter, hardens the egg.. 

But in spite of variations from case to case, there is one law to 
which there are no exceptions: every personality develops continually 
from the stage of infancy until death, and throughout this span it per- 
sists even though it changes. Each succeeding stage of development 
emerges, in very complex ways, from the stages existing previously. 
This process of transition from one stage to another is in some meas- 
ure clarified by our present knowledge. Though far from complete, 
this knowledge is extensive enough to demand our attention during 
the present chapter and the four succeeding. 


Human beings in common with all other living creatures are sub- 
ject to the laws of heredity. What these laws are, and to what degree 
they determine a man’s physique, temperament, mentality, and per- 
sonality are subjects of controversy. The force of this controversy 
can easily be seen by placing in opposition two extreme statements. 

Heredity and not environment is the chief maker of men. . . . 
Nearly all the misery and nearly all the happiness in the world are due 

1 Work of Art, by Sinclair Lewis, copyright, 1934, pp. 310L, reprinted with 
permission from Doubleday, Doran and Company, Inc. 



not to environment. . . . The differences among men are due to dif- 
ferences in the germ cells with which they are bom . 2 

With this statement a behaviorist’s notorious challenge may be com- 

Give me a dozen healthy infants, well formed, and my own speci- 
fied world to bring them up in and I’ll guarantee to take any one at 
random and train him to become any type of specialist I might select- 
doctor, lawyer, artist, merchant, chief, and yes, even beggar-man and 
thief, regardless of his talents, peculiarities, tendencies, abilities, voca- 
tions, and race of his ancestors . 8 There is no such thing as an in- 
heritance of capacity, talent, temperament, mental constitution and 
characteristics . 4 

But even such an extreme environmentalist as the author of the 
last quotation cannot and does not deny the inheritance of physical 
structure. If a child resembles his parents physically the fact is 
ascribed to the operation of determinants in the germ-plasm; but if a 
child has habits, handwriting, emotional outbursts, or esthetic sensi- 
bilities similar to his parents’, the author claims that training is suffi- 
cient to account for the resemblance. 

Yes, there are hereditable differences in form, in structure, but the 
mere presence of these structures tells us not one thing about func- 
tion . 3 

The behaviorists believe that there is nothing from within to 
develop. If you start with the right number of fingers and toes, eyes, 
and a few elementary movements that are present at birth, you do not 
need anything else in the way of raw material to make a man, be that 
man genius, a cultured gentleman, a rowdy or a thug.® 

Muscles, heart, glands, nerve-tissues, cortex, all are inherited as struc- 
ture, but so long as they are “normal” they do not predetermine 
function; psychology can therefore disregard them. 

This is a remarkable doctrine, one that asks us in effect to admit 
that gross (abnormal) defects of structure set limits upon function, 
and at the same time to deny that all other peculiarities of structure 
determine idiosyncrasies of function. The logic is amazingly faulty. 

2 A. E. Wiggam, The New Decalogue of Science, 1923, p. 42. 

8 J. B. Watson, Behaviorism, W. W. Norton & Co., 1925, p. 82. 

*lbid., pp. 74 f. 

5 Ibid., p. 77. 

6 J. B. Watson, Psychological Care of Infant and Child, W. W. Norton & Co. 

1928, p. 42. 


Even though glands, for example, be regarded as structure, their 
structural differences obviously produce differential fwnctions {viz., 
differences in temperament). Likewise, intelligence and adaptive re- 
flexes though closely bound to inherited nerve structure, also repre- 
sent an important -functional inheritance. 

Most psychologists tend whenever possible to stress the operation 
of environmental forces, even though they seldom state their case in 
as extreme a form as Watson, and even though they may not be 
behaviorists. There are two reasons for this. In the first place, the 
plasticity of the child, the rapidity of his learning, the subtle course 
of conditioning, the ease of “slanting,” give unlimited possibilities for 
the acquisition of peculiarly personal habits of response. With a nat- 
ural preference for causes that are apparent rather than for those that 
are hidden (and therefore only inaccurately inferred), the psycholo- 
gist in nearly every instance chooses to ascribe personal characteristics 
to conditioning, imitation, trauma, or some other form of learning. 
Since the possible variations of environmental influence are infinite in 
number, they can easily account for all the differences between 
human beings. Why invoke an explanation in terms of heredity? It 
often seems superfluous. 

Another matter seriously worries the psychologist. He knows that 
function invariably involves the integration of afferent, central, and 
efferent nervous pathways. In this circuit are many synapses, but 
synapses are non-cellular. They are indefinite junctions between neu- 
ronic cells. The laws of heredity, on the other hand, presuppose 
material determiners that reside in material substance. Does heredity 
also provide “tendencies” in the synaptic regions? If so, how? It 
would be far easier to regard only cell-structure as inherited and to 
ascribe all synaptic connections to the influence of learning or “neu- 
robiotaxis.” 7 This dilemma worries the psychologist, though for 
some reason it does not seem to worry the geneticist. One reads, for 
example, the words of Jennings: 

Temperament, mentality, behavior, personality— these things de- 
pend in manifold ways on the genes . 8 It may be safely said that there 
is no type of characteristics in which individuals may differ that has 
not been found to depend on genes . 9 

7 Even the reflexes that an infant displays at birth are, according to Holt, learned 
in utero. (E. B. Holt, Animal Drive and the Learning Process, 1931.) 

3 H. S. Jennings, The Biological Basis of Human Nature, W. W. Norton & Co, 

1930, p. 36. '“Ibid., p. 154. 



To illustrate the theory of genes, the evidence for which he claims to 
be “positive, inescapable, conclusive ,” 10 he refers continually to such 
functional and highly complex personal characteristics as laziness, 
stupidity, slowness, lack of industry, ambition, patience, and genius . 11 

Jennings, however, does not claim that these complex personal 
characteristics are genetic units, nor that they are determined ex- 
clusively by heredity. The manner in which genes play their part is 
in the highest degree complex. In the Drosophila at least fifty pairs 
of genes co-operate to produce the usual red color of the eye. 

By changing any one of the fifty genes of the fruit-fly that take 
part in producing the eye color, the color is altered; eyes of other 
color are produced; or there is no pigment in the eye; or it is struc- 
turally imperfect. The same situation is found for all characteristics, 
in the fruit-fly or in ourselves. Any feature or characteristic, struc- 
tural, physiological, or mental, can be changed or made defective by 
altering any one of the many different genes that co-operate to pro- 
duce it . 12 

A gene, we are told, is not a hypothetical unit with mystic properties, 
but an actually separable part of the elongated chromosomes . 13 Since 
genes enter into the production of every bodily cell it is not unthink- 
able that every structural characteristic receives its initial determina- 
tion from them. It is, to be sure, a leap from the determination of 
bodily structure to the determination of the functioning personality, 
but since function is undoubtedly intricately dependent upon struc- 
ture (Watson to the contrary notwithstanding), Jennings’ claim that 
heredity affects the traits of personality, though vague enough, is not 

This doctrine of genetic determination does not state that person- 
ality is inherited, but rather that no feature of personality is devoid of 
hereditary influences. It means simply that if the genes are altered 
the personal characteristics are altered, not that personal characteris- 
tics are determined solely by the genes. In fact, it admits that while 
every characteristic is influenced somehow by the genes, the same 
characteristic may be influenced also by the surrounding conditions, 
by temperature, by the physical environment, and in human beings, 
by the complex social environment. Even the young Amblystoma has 
the possibility of two diverse careers, as an aquatic or as a land animal, 
depending upon his early surroundings. Similarly, each human indi- 

10 Ibid., p. 36. 12 Ibid., p. 17. 


vidual so far as heredity is concerned has the possibility of many 
careers, and of many personalities, whose realization will depend upon 
the exigencies of his physical and social environments. 

There is then an embarrassing array of causes that may be drawn 
upon to explain a given personality. The gene theory provides by its 
flexibility more than enough permutations to account both for family 
resemblances and for individual distinctiveness. On the other hand 
the environment, in connection with the subtle operations of human 
learning, provides equally limitless possibilities. Since every quality 
is probably influenced by the original determinants inherent in the 
genetic system, and at the same time by the course of life in an 
actively stimulating environment, it becomes impossible to ascribe 
with finality any single feature of personality either to heredity or 
to experience. Both are always involved. This point of view might 
be expressed as an equation: Personality — f ( Heredity ) X (Environ- 
ment) . The two causal factors are not added together, but are inter- 
related as multiplier and multiplicand. If either were zero there could 
be no personality. 

The persistent reader may ask, which is on the whole more im- 
portant? For given individuals, or for given groups of individuals, the 
question is at least theoretically possible to answer. Within a family 
where very similar environmental influences exist for each member, 
the differences between siblings are probably due, paradoxical as it 
may seem, chiefly to heredity. Unless the children inherited quite 
different genes from the myriad combinations which their parents 
had to offer, they might be expected, because of the similarity of 
their environments, to be much more alike than they are. The same 
principle holds true in other homogeneous groups. For example, dull 
people, Jennings thinks, are on the whole determined relatively more 
by their heredity than bright people, for the latter learn with ease, 
adjust variously and with great delicacy to the environment, while 
the unintelligent plod along changelessly with such qualities and man- 
ners as spring from their original natures . 14 

The genes supply one set of conditions for development, the en- 
vironment another. In any given individual the environmental and 
genetic conditions may reinforce one another, to produce a trained 
and talented statesman, for example. Or they may conflict to produce 
a talented but untrained statesman, or one trained but untalented. In- 

14 H. S. Jennings, The Biological Bask o f Human Nature > 1930, p. 181. 


numerable combinations result, each affecting the structure of per- 

There is yet another complicating consideration: characteristics 
induced by genes may also be induced by environmental forces. (In 
personality, for example, one individual may be reclusive and retiring 
because of inborn qualities, another because of conflict with the en- 
vironment.) This consideration, if followed logically, means that one 
personality may be relatively more a product of heredity, and an- 
other of environment. In some cases the decision might be reached 
after an intensive analysis of the life-history; but it would be impos- 
sible to formulate a general rule regarding the weight of the two fac- 
tors in every life. In one case training seems markedly to outweigh 
the influence of heredity; in another the stress seems to be reversed. 

Besides the conclusion that in different lives the operation of 
heredity plays very different roles, there is another fairly certain fact 
important for the psychology of personality, namely, that the more 
directly a quality is bound to structural inheritance the less modifiable 
it is. The three principal raw materials of personality, physique, the 
endowment of intelligence, and temperament, are genetically deter- 
mined through structural inheritance, and are only slightly altered 
by conditions existing subsequent to birth. They are the effective 
agencies of heredity entering the process of growth at every stage to 
influence the development of traits and attitudes. Sometimes they 
accelerate the molding influence of the environment; sometimes they 
place limitations upon it; but always their force is felt. 


The newborn infant lacks personality, for he has not yet encoun- 
tered the world in which he must live, and has not developed the dis- 
tinctive modes of adjustment and mastery that will later comprise his 
personality. He is almost altogether a creature of heredity. In fact, 
if it were not for two complicating considerations we might equate 
the observable equipment of the neonate with the endowment of his 
inheritance. The complicating considerations are, first, the probable 
existence of pre-natal learning, and secondly, the fact that some 
aspects of inheritance are latent and require time for their maturation. 

Regarding pre-natal learning little needs to be said, for however 
many adaptive responses the child may learn in utero, he does not 
learn them in the environment in which he must live, and it is only 

Figure ii 

The Beginnings of Personality: The Situation at Birth 

SIMPLE SENSITIVITY (stimulation of extroceptive, introceptive, or propri- 
oceptive sense organs) 

SEGMENTAL DRIVE (tensions), e.g., hunger, thirst (cf. pp. 114-118) 
[GOAL-SEEKING PROCESSES: instincts, needs, entelechy (cf. pp. 112-1x4)] 


INNATE INDIVIDUAL EQUIPMENT : sensory capacities, nervous plastic- 
ity and retentivity (INTELLIGENCE); glandular and chemical tonicity 
(TEMPERAMENT); muscular and skeletal structure (PHYSIQUE); 
special capacities, skills, and sexual functions, to mature later 

ADAPTIVE MECHANISMS (potentialities for modification of the stream of 
activity), e.g., maturation, conditioning, integration, inhibition, differentia- 
tion, and all other agencies for modifying behavior during the course of 
the individual’s interaction with his environment (cf. Chapters V-VIII) 

MASS ACTION (Gross patterns of movement, often called “random 
activity”), e.g., extension, retraction, indeterminate squirming, accompany- 
ing vocalization, etc. 

SPECIFIC REFLEXES: breathing, sucking, swallowing, digestive and elimi- 
native sequences, grasping, sneezing, Babinski, etc. 

set into motion and sustained through 


(Vitalistic Postulate) (Mechanistic Postulate) 

reflected in 


differentiated at birth into 



with the distinctive adjustments to the post-natal world that the 
problem of personality is concerned. For our purposes, therefore, the 
congenital and the hereditary equipment of the child may both be 
regarded together as the one primordial source of personality, ob- 
servable, in part, in the stream of activity at birth. 

Less easily disposed of is the problem of maturation. Not every 
inherited tendency is observable at birth. Throughout life there is a 
subtle procession of ripening events, so much obscured by the effects 
of training that no one can tell just what develops in response to 
maturation and what comes from teaching. This problem will be con- 
sidered in some detail in the next chapter, but its existence makes it 
difficult for us to formulate an initial account of the “Givens” of per- 
sonality, as a starting point for a genetic theory of development. But 
since, in spite of the difficulty, a beginning must be made, an approxi- 
mate schedule of the equipment of the infant at birth is offered in 
Figure 11. 

The origin of the stream of activity here portrayed may be 
viewed either as a manifestation of a vitalistic urge ( Horrne , Will, 
Elan Vital), or of an equally mysterious though more scientific- 
sounding principle of “Protoplasmic Irritability.” Some Original 
Cause or Source of Animation is assumed explicitly or implicitly by 
all biological and psychological sciences. Life exists in individual 
forms. This fact, unexplained and perhaps inexplicable, is the starting 
point for these sciences. 

Recent studies of early infancy make a convenient distinction be- 
tween two manifestations of the observable stream of activity . 15 
There is, first of all, a type of diffuse, massive movement involving 
large regions of the body. Owing to the basic physiological opposi- 
tion of flexor and extensor muscle patterns (which have the original 
property of inhibiting each other while they are active) this mass 
action can frequently be characterized either as retraction (abience) 
or as extension (adience) ; while some of the random squirming that 
takes place, as in twisting, thrashing, pounding movements, makes 
such complex use of the musculature that it cannot readily be classi- 
fied either as retractive or extensive. 

From one study of the first ten days of life we are given the 
following picture: 

16 Cf. E. Dewey, Behavior Development in Infants, 1955. K. C. Pratt, “Specificity 
and Generalization of Behavior in New-born Infants: A Critique.” Psychol. Rev - 
1934,41,265-284. ' 

1 10 


The infant maintains continuous body movement with such speed 
and excessiveness that the experimenter, even when using a specially 
devised code, cannot keep up with the infant. The body squirms, 
twists, rolls, and bends. The back arches, the hips sway, and the head 
rolls from side to side or is thrown back. The arms slash vigorously 
and the legs are kicked in exaggerated extensor thrusts or are flexed 
sharply at ankle, knee and hip. Hands, feet, toes, and fingers are in 
continuous movement. Sucking and smacking sounds frequently occur, 
while loud crying is usually coincident with mass activity. All of this 
activity is more or less simultaneous and comes and goes with peri- 
odicities which appear in intervals of a .few seconds to several 
minutes. 16 

The same investigator reports that mass activity of this type is espe- 
cially prominent just previous to nursing, and occurs also from a 
flatulent condition, during regurgitation, intestinal disturbances, defe- 
cation and micturation. 17 That this diffuse and random movement is 
especially marked when there are ascertainable organic tensions is a 
fact of considerable importance for the theory of motivation (pp. 
113 ff-)* 

In addition to these gross patterns of movement, so difficult to 
characterize, there exist many specifically adaptive responses, such as 
movement of the eyes toward a light, sucking, swallowing, with- 
drawal of a limb from noxious stimuli. The exact number of these 
specific reflexes is difficult to determine owing to their frequent or- 
ganization into complex chains or into such large patterns or groups 
that their appearance approaches mass action itself. The operation 
of these reflexes is discussed in other texts. 18 Their importance for 
personality consists chiefly in the raw material they provide for the 
operation of two of the principal mechanisms of growth, viz., condi- 
tioning (pp. 151-153) and integration (pp. 138-147). 

What is it that sets the stream of activity into motion, that sus- 
tains it until it lapses or changes? This is the problem of motivation , 
and there is no problem in psychology more difficult to handle. The 
special portion of Figure 1 1, devoted to this topic, therefore will need 
much more thorough discussion. But before settling into it attention 
should be called to the two other “Givens” underlying the develop- 
ment of personality, 

16 O. C. Irwin, Genet. Psychol. Monog., 1930, 8, pp. 59 f. 

17 O. C. Irwin, loc. cit., and ]. Comp. Psychol., 1932, 14, 429-445. 

ls E.g., J. B. Watson, Behaviormn, 1924, pp. 90-103, and F. H. Allport, Social 
Psychology, 1924^ chap. iii. 



First, there is the innate individual equipment of physique, and 
temperament, and general intellectual capacity including the highly 
complex abilities and talents that later become manifest. Such consti- 
tutional endowment has many facets, and the part it plays in person- 
ality, as already indicated, is complex indeed. The question of tem- 
perament alone, intricately tied as it is to the chemical functions of 
the body, is as important as it is baffling. Take, as an example, that 
subtle constitutional quality commonly called energy, vitality, or 
“pep.” In various studies of leadership, of popularity, and of personal 
happiness this temperamental quality has been found to be decidedly 
significant. In common speech it is often equated with “personality.” 
What the precise physical counterpart of vigorous and vital activity 
may be one does not know, though certain glands, especially the 
hypophysis, the adrenals, the gonads and the thyroid are thought 
to have more to do with it than certain others . 18 

Or take the case of the normal speed of movement (likewise an 
aspect of temperament, or of one phase of temperament, viz., motil- 
ity). One investigator studied the normal rates of tapping of twins, 
of brothers and sisters, and of parents . 20 The results show a striking 
correspondence in normal tapping speeds among related individuals, 
monozygotic twins being closest, dizygotic twins and siblings less, 
parents and offspring next, while unrelated individuals had only a 
chance correspondence with one another. Such a finding is, of course, 
precisely what would be expected if heredity played an appreciable 
part in this basic form of activity, one of many of the tap-roots of 

The remaining “Given” in Figure 1 1 consists of adaptive mech- 
anisms that make it possible for the infant to vary his responses, to 
learn, and to bring his rich innate equipment into the most effective 
interaction with the demands of his environment. These mechanisms 
are not structures; they are merely instruments of growth, types of 
modifiability of which the nervous system is capable. It is capable of 
organizing its segmental responses into higher and more complex units 
through integration; of differentiating its diffuse, mass responses into 
more refined and successfully adaptive movements. It is capable of 
maturing, of being conditioned, of learning. In the following chap- 
ters all these instruments of growth will be considered in detail. 

19 Cf. P. Richter, Amer. J. Orthopsychiatry, 1932, 2, 345-354; D. J. Ingle, Psychol 
Rev., 1935, 42, 466-479. 

Ida F rischeisen-Kohler, Char, dr Pers., 1933, 1, 301-313. 

I 12 



In the central portion of Figure u there appears an entry in 
brackets: [Goal-seeking Processes: instincts, needs, entelechy ] . The 
brackets call attention to the fact that this “Given” is often regarded 
as a superfluous assumption. Psychologists dispute as to whether the 
equipment of the new-born infant does or does not include a system 
of latent purposes. Does the primordial stream of activity contain 
within itself directions which determine its own course of develop- 
ment? Is there in the infant, for example, a latent propensity that will 
later lead him to construct, and another to acquire, one to mate, an- 
other to imitate, another to seek companionship of his fellows? The 
instinct theory asserts that there are such propensities operating “prior 
to experience and independent of training.” An instinct “determines 
its possessor to perceive (to pay attention to) any given object of a 
certain class and to experience in its presence a certain emotional 
excitement and an impulse to action which finds expression in a spe- 
cific mode of behavior in relation to that object.” 21 Since it is ob- 
viously impossible to discern any such elaborately purposive disposi- 
tions in the conduct of the newborn infant, it has been necessary for 
proponents of the instinct doctrine to lean heavily on the theory of 
maturation which holds that this providential equipment of goal- 
seeking processes must take time to ripen, and that throughout life 
the instincts, one after another, come of age. 

In recent years it has become common to reject this somewhat 
extravagant portrayal of human purposes. It seems to many psycholo- 
gists to violate unnecessarily the requirement of parsimony in a scien- 
tific theory. Are there not many individuals who live their entire 
lives lacking some of these instincts, failing, for example, to become 
acquisitive, constructive, pugnacious, or parental in their behavior? 
Is it not simpler to account for these types of interest if and when 
they are present, than to assume that “instincts” are common to a 
species and then be forced to explain away the many exceptions 
where the “instincts” fail to put in an appearance? 

Yet in one form or another many contemporary theories of per- 
sonality are instinct-theories. They view personality as a matter of 
individual modification of universal instincts or common needs. A col- 
lector of vases is but showing a special modification or extension of 

21 W. McDougall, Outline of Psychology, 1923, p. no. 



his acquisitive instinct; an altruist is following the instincts of self- 
abasement and gregariousness (perhaps with a little of the parental 
propensity thrown in). But are not the purposes of different people 
far too diverse and too numerous to be traced to a few primal motives 
shared by all the species? Are the directions of striving after all in- 
nately determined? Is it not necessary to allow for the learning of 
new motives and for the acquisition of novel interests as personality 
matures? When people do seek the same goals may the fact not be 
explained by the more parsimonious assumption that similarly con- 
structed individuals living in similar environment, influenced by simi- 
lar culture, 'would develop similar goals and employ similar modes of 
obtaining them? 

If so, then instincts would evaporate. They would turn out to be 
nothing more than constellations of emotion, habit, and foresight^ 
better called sentiments or interests , and regarded as acquired rather 
than innate. Learning would then serve not merely as a way of ex- 
tending and modifying purposes, but also of creating them. And the 
fact that many of these purposes are fairly common to mankind could 
be readily explained without recourse to a hypothesis of racial in- 

The question of instincts in animals should not confuse the issue, 
for there is such flexibility in learning, in making and breaking habits, 
and so much insight, foresight, and delay in response, that human 
goals must be viewed as different in kind from the stereotyped ob- 
jectives of lower animals. 

Rejecting the instinct hypothesis it becomes necessary to provide 
an alternative. A beginning, but only a beginning, may be made in the 
theory of Drive. A drive is defined as a vital impulse which leads to 
the reduction of some segmental organic tension. It has its origin in 
an internal organic stimulus of peculiar persistence, growing charac- 
teristically stronger until the organism acts in such a way as to al- 
leviate the accumulating tension. 

The doctrine of drive is a rather crude biological conception often 
employed as a factotum by psychologists with a simple mechanistic 
outlook. The hypothesis herewith offered is that the doctrine, while 
inadequate to account for adult motivation, does none the less offer a 
suitable portrayal of the motives of young infants, and for that 
reason serves very well as the starting point for a theory of motiva- 
tion. After the level of infancy is passed primitive segmental drive 
rapidly recedes in importance, being supplanted by the more sophisti- 


cated type of motives characteristic of the mature personality, and 
commonly represented by such terms as interest, sentiment, value, 
trait, ambition, attitude, taste, and inclination. Obviously none of 
these motives are found full-fledged in the new-born child. 


The doctrine of drive must be set where it belongs in a well- 
saturated background of biological and evolutionary theory. This 
theory, readily intelligible to all readers of biological science, may be 
presented in a single sentence. 22 The personality of an individual is 
the mode of adjustment or survival that results from the interaction 
of his organic cravings {segmental drives) -with cm environment both 
friendly and hostile to these cravings , through the intermediation of 
a plastic and modifiable central nervous system. 

Just as all organisms take on the form of some species, each of 
which represents a successful mode of survival in the evolutionary 
struggle, so too do individuals within the human species attain per- 
sonality as the form of survival most suitable to their individual needs 
within the particular environmental framework provided. The central 
nervous system in the process of effecting the necessary adjustments 
between the organic cravings and the exigencies of the environment 
develops certain characteristic habits, attitudes, personal traits, forms 
of sublimation and thought, and it is these characteristic modes of 
adjustment that, taken collectively, comprise personality. In a sense, 
therefore, the central nervous system is the seat of personality. 

The organic cravings which are the sole original motive power 
for activity are closely associated with the primitive vegetative proc- 
esses having their origin in various segments of the autonomic 
nervous system. When the vital functions of the viscera are disturbed 
widespread postural changes result throughout the body, and the 
organism is restless and active until an equilibrium is reached. The 
nutritional system illustrates the situation very nicely. When the 
stomach is empty, its contractions bring into play glandular, cardiac 
and respiratory changes which with their consequent effect upon the 
musculature of the body create a condition of restlessness and ran- 
dom activity that only the ingestion of food can appease. The origi- 

22 The exposition of the biological theory given in these pages follows quite 
closely the formulation of E. J. Kempf, in “The Autonomic Functions and the Per- 
sonality/’ Nerv, & Ment, Dis* Monog, Ser. % 1921, No. 28. 



nai sucking and swallowing reflexes are nature’s only provision at the 
outset for meeting this type of maladjustment; but as the org anis m 
grows, the central nervous system evolves elaborate new ways of 
effecting the adjustment. The individual learns to use language (to 
ask for food), to earn money (to buy food), to organize markets for 
barter and for efficient distribution, and to obey certain social re- 
quirements in the preparation and eating of food. All these elabora- 
tions are ultimately in order that the organic craving may be most 
efficiently satisfied within the restraining conditions of the environ- 

The vegetative nervous system, where these cravings originate, 
is thought to be more primitive and more essential than the central 
nervous system which is primarily an agency of adjustment. The 
vegetative system is the master, the cerebro-spinal is the servant; the 
former compels adjustment, the latter effects adjustment. 

Evidence for the primacy of the vegetative system might be 
drawn from various sources. (1) Kempf, for example, points low in 
the evolutionary scale (to the amoeba, for instance) where the vege- 
tative functions are all essentially present, but without a sensori- 
motor system to mediate between them and the environment. The 
vegetative system is directly submerged in environmental media and 
supplied immediately with the requirements for effective survival. 
(2) Obviously, too, the vegetative functions of the infant are far 
more advanced in development than those of the higher nervous 
system. Indeed a child with only seven months of pre-natal life has 
a complete and complex autonomic development with only a few as 
yet undeveloped central nervous functions to serve it. (3) In some 
individuals the secondary nervous system never grows, and yet the 
individual (an idiot) may live a long time, provided only that some- 
one else supplies the idiot’s autonomic system with its requirements. 
(4) In older people the secondary system may disintegrate entirely 
as in dementia, and yet the individual will live so long as his vegeta- 
tive functions are not impaired, and so long as others take over the 
role of his own deteriorated intellectual functions. (5) The study of 
the effects of fasting, of brain lesions, and of accidents to the cortex, 
all confirm the view that life depends not upon the integrity of the 
functions of the central nervous system, but on the intactness of the 
autonomic segments and their vital nervous centers. (6) In the case 
of urgent organic cravings their satisfaction takes precedence over 
all forms of activity. The central nervous system is compelled to 


serve them. Lack of oxygen, intense hunger, marked sexual tension, 
the need for elimination, may all defeat the “intellectual” activity of 
the central nervous system and compel it to find ways of satisfying 
the urgent craving. This list of arguments in defense of the primacy 
of the autonomic system reads very much like Schopenhauer’s de- 
fense of the primacy of the Will over the Intellect . 23 It is an anti- 
rationalistic view of human activity that today finds much favor . 24 

Figure 12 indicates schematically the essential features of this 
theory. It shows that Drive is ultimately of autonomic origin, that 
the environment supplies both potential satisfactions as well as 
menaces and obstacles to the well-being of the organic system. The 
central nervous system receives warning of both disturbing and satis- 
fying elements in the environment, and in general operates in such 
a way as to acquire the maximum of gratification for the affective 
cravings with a minimum expenditure of energy. Strictly speaking, 
of course, the central nervous system and the autonomic are not 
entirely independent; they constantly exchange impulses and elab- 
orately interact. What the theory does is to call attention to their 
supplementary roles in adjustment, one as motivational, the other as 

According to this theory the most important organic cravings 
(drives) are the nutritional and sexual. Of these two the latter is of 
cardinal significance for personality, for in civilization the obstacles 
to the free and natural satisfaction of this craving are greater than 
those confronting the nutritional drive. The chief single source of 
personality is, therefore, the unsatisfied sexual craving that provokes 
continuous and circuitous efforts within the central nervous system 
to invent some adequate means of resolving the erotic tensions. The 
force of this affective craving, as is now well known, is the root of 
many neurotic adjustments that represent unconsciously artful but 
still futile attempts of the central nervous system to supply a solu- 
tion for insoluble personal problems. It is not always apparent which 
personal characteristics are outgrowths of attempts to solve personal 
sexual problems, but to a trained ( e.g ., a psychoanalytic) eye, a 
manner of talking, flushing, working, saving, day-dreaming, or phi- 
losophizing may reveal the struggle. 

The sexual tension, important as it is, is not the only source of 

23 A. Schopenhauer, World as Will and Idea, II, Appendix, chap. xix. 

24 Other authors besides Schopenhauer who are responsible for the prevailing 
voluntaristic or anti-rational view of motivation are Darwin, McDougall, and Freud 



traits. The doctrine of Drive recognizes others. In the case, for ex- 
ample, of a fear that besets an individual when confronted with some 
continuous menace from the environment, such as ridicule for some 
physical weakness— in such a case there result all manner of distinctive 

Autonomic Nervous — > Central Nervous — » The 

System < — . System < — Environment 

segmental cravings, 
both avertive and 
acquisitive, . caused 
directly oE indi- 
rectly by the oper- 
ation and interac- 
tion of organs in- 
nervated by this 

pituitary gland 
pineal gland 
lachrymal gland 
dilator of pupil 
salivary gland 
hair follicles 
sweat glands 
thyroid gland 
thymus gland 




visceral arteries 




adrenal gland 







unlearned cerebro- 
spinal reflexes 
conditioned reflexes 
habits and skills 

expressive movements 
expressive postures 

philosophy of life 
neurotic adjustments: 







all other ingredients 

Figure 12 

potential satisfaction 
(e.g. y food, love, ob- 
jects, oxygen, sun- 
shine, water, and 
objects of culti- 
vated interest) 
potential menace 
(painful and inju- 
rious stimuli, threat- 
ening and destruc- 
tive objects) 
obstacles to the secur- 
ing of satisfaction 
or the avoiding of 

Biological Theory of Personality (Kempf) 


compensations and escapes to avoid the painful tensions (inferiority 
feelings) that result. The frail boy who suffers a “narcissistic wound” 
every time he is taunted by his fellows, and finds refuge in becoming 
a teacher’s pet; the short man who cultivates a beard, grand gestures, 
and a haughty bearing; the man who is excessively polite in order to 
escape abuse or domination by others; the scrupulous monk attempt- 
ing to expiate persistent feelings of guilt: all are cases of central 
nervous adjustments (and forms of personality) resulting from per- 
sistent organic tensions involving fear. Personality results from the 
attempts of the central nervous system to establish security and com- 
fort for the individual torn between his own affective cravings and 
the harsh demands of his environment. 

To the objection that man’s cravings are after all not exclusively 
nutritional, sexual, fearful, or otherwise “vegetative,” the biological 
psychologists (Kempf, for example) reply with the behavioristic 
theory of conditioning. A craving may be reinstated easily by vari- 
ous originally indifferent stimuli which have acquired potency 
through their association with the effective stimulus. Take the case 
of attachment to one’s parents, the so-called filial sentiment. At first 
the infant’s cravings are only for bodily comfort. The mother be- 
comes associated with this comfort; later her mere presence is suffi- 
cient to bring pleasure and her absence provokes a longing for the 
comfort of her companionship. Such extension and conditioning may 
continue until objects, tastes, ideas, associated with the mother be- 
come satisfying when they are present, and set up desire and longing 
when they are absent. Due to such conditioning maternal ideas are 
accepted with emotional conviction, and violation of parental teach- 
ing even in later life brings restlessness, pangs of conscience, and 
feelings of guilt. 

The theory goes on to a treatment of mental conflict. Basically 
all drives are either acquisitive or avertive. One wishes to have some- 
thing that the environment affords or to avoid something in the en- 
vironment that threatens one’s health and safety. This opposition 
leads to many types of conflict, for in the course of life the desirable 
is often tangled with the undesirable; the brightness of the flame 
attracts but its heat repels; danger is often alluring. Intelligence (a 
functional capacity of the central nervous system) is sometimes able 
to discriminate successfully and to attain the desirable while exclud- 
ing the undesirable. But often the conflict is not intelligently resolved, 


1 19 

and a suppression of the desire occurs because of the hazards attend- 
ing its realization. Or the object of desire is obtained bringing with 
it remorse and regret because of the attendant evil consequences. 
Sometimes an alternation takes place; the individual swings back and 
forth from the self-indulgent to the ascetic course of conduct. Again, 
there may be a compromise, whereby a little of the desirable and a 
little of the undesirable are both admitted. Which method of solving 
conflicts the individual characteristically takes may be an important 
feature of his personality. Some people are nearly always headstrong, 
others hesitating and undecided, still others adopt the road of sup- 
pression, and some seek always the golden mean . 25 

Only in its broader outline is the biological theory described 
above acceptable. Personality, let it be admitted, represents the mode 
of survival that the individual has consciously or unconsciously 
worked out for himself. It is a resultant in a parallelogram of forces. 
But the forces are not quite so simple as the narrower statement of 
the biological theory would imply; for one thing the personality it- 
self supplies many of the forces to which it must adjust! 

To make the criticism more specific, there is first of all a tendency 
to keep the list of primary organic cravings too limited. If we con- 
cede first place among the biological motives to the nutritional and 
sexual tensions, there are still other original needs— for example, ten- 
sions demanding sleep, equalization of temperature, elimination, the 
unimpeded freedom of each organ and limb— all of which may under- 
lie the development of important personal characteristics. In children 
there is clearly a craving for almost continuous muscular exercise 
during waking hours. Parents learn, often to their astonishment, that 
this vital need (not corresponding to any autonomic segment) takes 
precedence over nutritional demands; even when hunger is present, 
children frequently prefer physical activity in play to eating. 

Further, the division of labor between the two nervous systems 
(in Kempf’s theory)— one as master and one as servant, one as motive 
and one as agent— is far too simple. It is known for example that 
cortical determination plays a decisive role in the sexual behavior 
even of lower animals, and most obviously, of course, in the sexual 
and all other motives of men . 28 As soon as the earliest stages of 

25 Cf. E. B. Holt, The Freudian Wish, 1915. 

26 Cf. K. S. Lashiey, Psychol. Rev., 1924, 31, 192-202; also J. F. Fulton, Muscular 
Contraction and the Reflex Control of Movement , 1926, chap. xxi. 



infancy are past every autonomic tension is overlaid with intellectual 
and volitional factors that facilitate, inhibit, or otherwise direct the 
course of the motive. Hence, the compartmentalization of the auto- 
nomic and cerebro-spinal nervous systems cannot be rigidly main- 

Current emphasis upon “glands regulating personality” is an 
article of faith wholly in keeping with an over-simplified biological 
theory of personality, but it is psychologically far too narrow a 
conception. Granting that the chemistry of the body has much to do 
with the general cast of temperament, and that severe dysfunctions 
of the glands bring with them characteristic types of emotional dis- 
order, there is still no reason to suppose that a specific and pro- 
portionate relationship exists between the chemical and psychic con- 
stitution of normal people. Tensions produced by glandular activity 
are absorbed into the more integral tensions that comprise personal 
motives. Suppose, for example, that there is a marked secretion from 
the adrenal glands. A vague emotional excitement will probably 
ensue; but the way in which this excitement is handled is a matter 
of deep-seated habits and attitudes, even of one’s underlying philoso- 
phy of life. Chemical changes induced by disease or age, the meno- 
pause, adolescence, even hyperthyroidism and castration, are all 
handled by the individual in ways characteristic of his own pre- 
existing personality . 27 

A final warning should be given against the acceptance of a 
narrow biological theory, this time in respect to its superficial con- 

27 Endocrine enthusiasts commit their fallacy of exaggeration chiefly because 
they do not distinguish between personality and temperament . They do not realize 
that there is a great difference between simple temperamental correlates of endocrine 
action (excitability, irritability, apathy, fluctuation of mood, etc.) and the much 
more intricate, cortically dominated, traits of personality (e.g., egotism, aestheticism, 
pride, and suspiciousness). 

A much-needed antidote to the exuberance of the endocrinologists is a study 
based on 1,400 autopsies by W, Freeman. Using the method of comparing the size 
of various endocrine glands (assuming that abnormal size indicates some variation in 
function) with the records of the patients' personalities, he reaches the following 
conclusion: “As far as determining whether an individual shall be a proud, sensitive, 
suspicious, paranoid individual or a timid, shut-in dreamy, schizoid person; a bois- 
terous, jolly, hail-fellow-well-met cycloid, or a moody, pedantic, egocentric, epilep- 
toid individual, the endocrine glands would seem to have little say in the matter.” 
Annals of Intern. Med., 9, 1935, pp. 444-450. 

A general survey of the findings of endocrinology in relation to personality is 
offered by D. J. Ingle, Psychol. Rev., 1935, 42, 466-479. Normally, this author points 
out, the endocrines are concerned with maintaining homeostasis, and by themselves 
lack potency for bringing about significant differences in personality. 



ception of learning. It is all too easy to say that the primitive organic 
tensions become conditioned, and by this bit of verbal magic to think 
that one has accounted for all the motives of the adult person whose 
desires include not only nutritional and sexual satisfactions but like- 
wise fine music, rare books, and the answer to puzzling problems in 
science, politics, and theology. What really happens is that an elabo- 
rate process of learning and growth intervenes between the organic 
wants of infancy and the cultural wants of adulthood, involving all 
manner of linguistic, imaginal, and rational factors that ultimately 
transform the segmental cravings of infancy into desires having no 
longer any functional connection with them, but holding in their 
own right an autonomous place in personal life (cf. Chapter VII). 

The biological theory, then, is at its best when applied to the 
simpler motives of infancy. It is right in holding that there is no 
need to postulate an elaborate system of latent purposes underlying 
the restless behavior of the young child. On the other hand, as per- 
sonality develops it is not possible to consider the cruder segmental 
motives of infancy as all-sufficient. Even though the adult, no less 
than the child, fulfills the requirements of survival, and fashions 
habits and attitudes that relate him adaptively to the world in which 
he must live, still the instruments he uses, and the complex roles he 
fills, place him beyond the reach of the cruder biological conception 
wherein segmental drive and conditioning are regarded as the simple 
and sovereign sources of personality. 


Returning to the infant: the evidence then tells us that at birth 
he is endowed with capacities for activity of both a random-diffuse 
nature and of a specific-reflex order; these capacities can be activated 
by various forms of stimulation, and especially by tensions arising 
in certain autonomic-somatic segments that provoke restless move- 
ment until the tensions are resolved. He is endowed also with a dis- 
tinctive physical and nervous constitution, including peculiarities of 
glandular function that predetermine to a large extent both his tem- 
perament and course of physical growth. Also in his constitutional 
inheritance are various incipient talents and defects that will in time 
manifest themselves, given suitable environmental conditions. In addi- 
tion to this lavish equipment the infant has a many-sided capacity 



for mental growth involving numerous neuropsychic mechanisms 
that later must be considered in detail. 

But in spite of all this equipment the newborn infant lacks per- 
sonality. Although many of its determinants are congenital, person- 
ality as such is not inherited. Only when the original stream of 
activity meets the environment, acting upon it and being acted upon 
by it, do the first habits, conscious desires, and incipient traits emerge. 

To discover the earliest manifestations of personality, two psycho- 
logically-minded parents observed a young infant attentively during 
the first months of life, keeping a complete daily diary. Oddly 
enough, for all their watchfulness, there are no certain entries con- 
cerning personality in their child until the fourth month of life. It 
is, of course, true that an infant in isolation cannot be viewed in re- 
spect to his individuality as easily as if he were in a group of children 
(or at least with one other child) of his own age. Some observers 
(able to compare the behavior of several infants) claim that the first 
faint signs of distinctive adjustment appear soon after birth. 

The infant for whom the diary was kept will be called Andrew. 
A few excerpts reported here illustrate the difficulties of interpreting 
early behavior as expressive of personality. The age of the child in 
months and days is given before each entry. 

o. i First seen at age of 45 minutes, A. has his eyes open; much ran- 
dom squirming, clasping and unclasping hands. At the age of six 
hours, eyes wide open, blinks, yawns, and sneezes, also rejects 
first feeding from bottle. Aged 20 hours brought for first nurs- 
ing, A. is wide awake and clearly sucking upon his arrival, show- 
ing many facial and mouth reflexes; suddenly goes sound asleep 
and cannot be roused for nursing. 

0.2 Second time he is presented to breast A. is awake; he performs 
with perfection, sucking and swallowing, holding all the while 
tenaciously to the nipple. 28 Sucking movements take place against 
blanket or other objects, or frequently in the absence of any 
contact. 29 A. turns his head in well-co-ordinated fashion from 
right to left and back again. Coughs and hiccoughs noted. 

28 Inasmuch as A. nurses lying on his side, there is no evidence for the theory 
that swallowing is a conditioned response that depends upon the operation of grav- 
ity for the first stimulation of the successive ring muscles. Swallowing is entirely 

- 9 Some behaviorists claim that the sucking response takes place only upon 
actual contact with breast or other objects. This entry illustrates the existence of 
spontaneous (internally provoked) adaptive behavior, and warns against the familiar 
bias of behaviorists that all behavior is purely reactive. 



Already there is dear evidence both of the random mass-action and 
of the specific adaptive reflexes. 

0.3 Foot and leg reflexes noted: when sole of foot is tickled, all five 
toes contract together, and again only the four little ones (posi- 
tive Babinski). Prick on sole causes immediate withdrawing at 

04 Can be wakened by tickle but not by more massive movements. 
Grasp reflex noted in position of hands: thumbs inside. Eyes 
move synchronously, not independently. 30 

0.5 Seems to focus eyes on finger held 12 inches from his face, but 
does not follow the movement of the finger with his eyes. 

0.6 Startles at sudden (but not loud) sound of electric light switch. 

0.8 Quite definite fixation of eyes upon mother’s face, and upon 
flower held in front of own face. After fixating objects for a 
short time there is much synchronous roving of the eyes. 

0.10 When cheek is stroked lightly, muscles twitch in pattern of 
smile; when repeated a third and fourth time, a more diffuse 
bodily movement takes place, effecting a change of position and 

0.12 When pushed by his shoulders away from breast A. shows active 
resistance, a contrary movement of body (specifically at shoul- 
ders) back toward breast. 

0.14 As soon as he is lifted a little way toward the breast, but before 
he reaches it, A. stops crying, and makes vigorous sucking move- 

This entry at two weeks is the first observed instance of possible con- 
ditioning. Andrew’s preparatory adjustment for feeding seems to be 
the first visible example of learning. Thus far the entries have shown 
the dominance of sleep, random activity, and adaptive reflexes. 
Now that learning enters the record the possibility of distinctive 
modes of adjustment becomes enhanced. Up to this time there is 
little to distinguish Andrew from all his infant contemporaries. 

0.16 Mother steps out to the baby-wagon to claim A. for his feeding. 
She approaches one sleeping baby (presumably A.): nurse says, 
“That’s not your baby; that’s Baby S.” Mother next misclaims 
another sleeping baby as hers. In physical appearance the three are 
similar. Upon close inspection it seems that the chief visible dif- 

30 In some children this co-ordination seems to be acquired; in others it is pres* 
ent from birth. 



ference between A. and the others is that when crying he does 
not, as they do, display his gums. 

This is the first entry of “distinctive modes of adjustment.” In the 
case of Andrew the first trace of personality seems to be a matter 
of gums! 

The earliest entries bearing unequivocally upon personality come 
at the ages of three and four months. 

3.01 A notable feature of A.’s disposition is the relatively small amount 
of crying and large amount of smiling. He is easily amused by 
sounds, grimaces, wagging fingers, playful thrusts, etc. 

4.00 Characteristics at four months: healthy, good-natured, smiles 
readily and is easily amused, coy (while nursing he withdraws, 
smiles at mother, aggressively returns to breast with a kind of 
divided attention that can only be described as coy). Prophecy 
based on present characteristics: ready laugher, well-adjusted, 
i.e., “normal” and “extroverted,” capable of considerable temper, 
active, sensitive to rhythm, adaptable, wiry and muscular, tall, 
mischievous, with linguistic superiority. 

It will be noted that some of the entries at 4.00 are in the nature of 
prophecies. The qualities in question could not actually be observed, 
but there seemed to be faint indications that suggested the begin- 
nings of such qualities. It is not certain whether some of these pre- 
dictions would not be duplicated for my infant of the same age; 
nor is it certain that the parents are not prophesying wishfully or 
reading into the child parental qualities. Let us follow this prospect 

These prophecies of the parents suggested an experiment, namely, 
to check them against later records of Andrew’s personality. During 
the early years the parents themselves made the “follow-up” record, 
but after four years of age different teachers co-operated without 
any knowledge whatsoever of the entries made by the parents or by 
the other teachers. Figure 13 gives the results of this comparative 
study at different stages in early childhood. 

According to this table the prognosis at the age of four months 
is borne out in most respects. “Temper” seems to have given way to 
excitability, and “mischievousness” to imaginativeness. Thus two of 
the initially dominant characteristics have shifted their emphasis. Ap- 
parently a reserve, especially in school, has entered his nature, making 
him more reticent and less dominant socially. But on the whole the 



schedule is consistent throughout. Even at the age of nine, Andrew 
does not show any major characteristic not indicated in this chart. 
He is essentially a bright, not rugged, excitable child with marked 
masculine interests, adaptable to groups and friendly. 

From this record two important hypotheses concerning person- 
ality in early life are suggested. In the first place, it appears that 
vague and variable indications of distinctive traits are evident at an 
early age , in this case at four months. The second hypothesis is of 
special and far-reaching significance: from early infancy there is con- 
sistency in the development of personality . These principles, which 
the following pages will further establish, show that innate determi- 
nants of personality are indeed important. They do not, however, 
imply that personality is nothing more than an unfolding of latent 
tendencies uninfluenced by environmental circumstance. As Andrew’s 
environment becomes more diverse and more hostile, his assertiveness 
may be modified, or even destroyed, and his former “extroversion” 
may turn into a more marked withdrawing and reserve; conceivably 
even his gay and excitable temperament may undergo alteration. One 
can say only that he seems to have developed certain qualities com- 
patible with his own temperament, but that these have been, and will 
continue to be, either accentuated or diminished according to the re- 
quirements of his environment. 

There have been many other biographical studies of infants, but 
most of these have been concerned with the development of sensory, 
motor, and intellectual functions. The observer, especially if he is 
also the parent, is often inclined to overestimate the child’s abilities, 
or to interpret the behavior in the light of his own adult preconcep- 
tions . 31 But even with these limitations the intensive study of the 
single child has its advantages over the present more popular method 
of an impersonal observation of masses of children, the subtleties of 
whose natures cannot but be lost in a maze of mechanical methods 
for classifying, recording and averaging. There are, however, a few 
studies of the early development of personality with which the case 
of Andrew might be compared. 

Some observers have reported marked differences in the amount 
of activity and in the expenditure of energy not only in the first 
weeks but even in intra-uterine life. From the very beginning some 

S1 M. C. Jones has shown that 8y per cent of these biographical studies report 
development more rapid than the norms established for a large representative group 
of infants. Fed . Sem., 1925, 33, 537-585. 


Prognosis at 

4 mos . 

( parents ) 
ready, easy 

Record at 

12 mos. 
laughs readily in 

Record at 

2 yrs j? mos. 

( parents ) 
happy and gay 

Record at 

4 yrs., 7 mos. 
(teacher i) 

well adjusted 



sociable but not 

plays easily with 
other children 

likes to be with 
older children 


aggressive in 
friendships with 
other children 

dominates other 
children; demands 
to be admitted to 
groups of older 

leads and dominates; 
has to be put with 
older children 

active; capable of 

excitable; some* 
what nervous 

decidedly excitable 

very excitable, too 
tense; relaxation 
difficult for him; 
excitement may 
block speech 

sensitive to rhythm 

outstanding in 
school in rhythm 


regards people as 
agents for play; 

especially enjoys 
visitors; natural 
with strangers 

discusses freely; 
good listener 


many inventions 
of games 

people comment 
on his imaginative 
nature more than 
on any other quality 

much originality 
in drawing 


gets ideas quickly 



Comparative Records of One Child’s 

babies seem placid and contented; others fidgety, restless and enter- 
prising. Specifically, different patterns of feeding can be noted ( e.g ., 
the aggressive, intermittent, indolent, or impatient), and these pat- 
terns some writers believe are prophetic of a child’s later tempera- 
ment. 32 

One observer claims that there are differences in “social re- 
sponsiveness” as early as the twentieth day, 33 but in this study social 
responsiveness must be taken merely to mean temperamental differ- 

32 Cf. S. and M. Blanton, Child Guidance, 1927, p. 31. 

23 H, Zoepffel, Z sch. f. Psychol., 1929, in, 273-306. 


Record at 

6 yrs. 

(; teacher 2) 

almost always happy and 
active; easily provoked to 

Record at 

8 yrs . 

(teacher 3) 

unusual sense of humor; 
average amount of 

Record at 

9 yrs. 

(; teacher 4) 

without complexes or con- 
diets; no shyness or shame 

seems sensitive but never 
inattentive to what is 
going on 

well-adjusted, “sane”; has 
no enemies 

exceedingly sociable; out- 
standing in any group 

average in activity 

plays all games vigorously, 
prefers active boys of 
masculine type as friends 

jumps up and down 
when excited; occasional 
twitching in face; must 
be protected from exciting 

calm and well-mannered 
at school (parents’ note: 
at home entry at 6 yrs. 
by teacher 2 holds) 

likes to sing, so much 
that he must be silenced 
in school for drowning 
out other children 

attentive and interested 
in music 

of all class members uses 
strongest colors and sharp- 
est contours in drawing 

his playmates are older; 
gets along well with them 

fits into any group though 
with reserve 

fits well into group situa- 
tions at school 

imaginative in play; very 
fond of stories; likes 
pranks and tricks 

markedly original in con- 
tent of drawings; amusing 
detail; not mischievous in 

high in school work 

high in school work 

scholarship of high rank— 
I.Q. 140 


Personality During Nine Years 

ences in alertness and motility upon which differential social traits 
may later be founded. No distinctive social behavior as such can be 
observed as early as the twentieth day. But in the second half-year 
there are unmistakable differences in sociability. Biihler finds that 
some babies pay very little attention to people, whereas some are 
socially dependent, and still others are neither inattentive nor unduly 
dependent, but have what might be designated as “social ease.” 34 It 
should be noted that, as in the case of Andrew, these earliest observa- 

34 Ch. Biihler, “The Social Behavior of the Child,” Handbook of Child Psy- 
chology, 1st ed., 1931, chap. xii. 



tions establish distinctiveness in temperament, motility, and intelli- 
gence. Traits, interests, tastes, social habits, and sentiments are un- 

The second hypothesis, namely, that early observed differences 
tend to persist, likewise secures confirmation. An investigation by 
Shirley following the activity of twenty-five infants for two years 
from the time of birth, showed that certain qualities (e.g., adapta- 
bility, timidity, aggressiveness) seem outstanding in a child from the 
early months onward. 35 

In another study it was found that infants between the ages of 
two and twelve months could readily be classified as those who 
laughed more often than they cried, those who laughed and cried in 
approximately equal amounts, and those who cried more than they 
laughed. This temperamental characteristic was found to be generally 
constant throughout the early months of life. 36 It would be valuable 
if such a study were continued beyond the first year to see whether 
this early affective balance is a characteristic of temperament in later 
childhood and in maturity. 

One more study may be added to show that temperamental differ- 
ences are distinguishable in the early weeks of life, and that they 
persist throughout infancy. 37 Monozygotic twins were chosen, strik- 
ingly alike in appearance, anthropometric measurements, hand prints, 
and intelligence. They were alike also in certain basic forms of emo- 
tional expression, gestures of request and refusal, vocalization, good 
nature and amenable demeanor. But throughout the study T was 
more approachable, more aggressive socially, more fearless, more 
vivid and reactive. By contrast C was more immobile, more intro- 
verted, cautious and fearful. A few selections from the examiners’ 
protocol show the persistence of these differences. 

Age in 


o T more active immediately after birth; weight 3 oz. more than C. 

6 T more responsive to material, but fusses more readily than C. 

8 T more alert in fixating on objects; more reactive in prone posi- 

1 6 T cries more easily but also smiles more readily. 

24 T slightly more placid during examination. 

28 T vocalizes more; shows more tendency to fuss; more active. 

36 M. M. Shirley, The First Two Years, Vol. Ill, 1933, p. 219. 

50 R- W. Washburn, Genet. Psychol. Monog., 1929, 6 , Nos. j-6. 

37 A. Gesell and H. Thompson, Genet. Psychol. Monog., 1929, 6, No. 1. 


Age in 

3 6 T’s emotional behavior more spontaneous and vivid. 

38 T gives more heed to C’s material than C does to T’s. 

44 Vocalizations slightly more in C. Social playfulness before mirror 
slightly more vivid in T. 

62 T more active. Slight personality difference; extensive correspond- 
ence remains. 

79 T more mobile and active, but in adaptive, emotional and lan- 
guage behavior resemblance is extensive. . 

With the exception of one or two observations (at 24 and 44 weeks) 
T was the more reactive twin on every occasion. It is probable that 
the slightly different patterns of emotional nature will result in the 
development of increasingly distinctive personalities as soon as the 
social environment becomes broadened and as soon as the twins be- 
come aware of their similarities and differences. 38 

Only one study begun after the first year of life will be men- 
tioned. It is based upon observation of 72 children between two and 
three years of age. 89 Two traits, negativism and distractibility, were 
chosen for study. Ratings were made by different observers who 
spent much time in watching the children. After an interval of nine 
weeks the ratings for these children correlated .80 for negativism 
and .83 for distractibility. After an interval of forty-three weeks the 
ratings for these children correlated .60 for negativism and .64 for 
distractibility. Since these correlations are based on quite independent 
ratings by different observers, the correspondence is undoubtedly 
significant. These two traits, then, seem to function persistently 
throughout the age-range studied. Attention should be called to the 
fact that these traits are primarily of the temperament-motility type. 

From the evidence now in hand four important conclusions may 
be drawn: ( 1 ) Personality, defined as the distinctive mode of adjust- 
ment adopted by each individual in his efforts to live, is not formed 
at birth, but may be said to begin at birth. (2) The earliest distinc- 
tive adjustments in respect to which infants can be said to differ are 
in the intensity and frequency of their spontaneous activity (mo- 
tility) and in their emotional expression ( temperament ). Both these 
factors are primarily products of inheritance. (3) Probably not before 
the fourth month is there sufficient learning and maturation to form 
distinctive habits of adjustment or rudimentary traits; but by the 

38 Cf. H. von Bracken, “Mutual Intimacy in Twins,” Char, it Pers., 1934, 2 , 
2 93 - 3 ° 9 - 

*®F. L. Goodenough, J. ]uv. Res., 1929, 13, 204-219. 



second-half of the first year adaptive responses to the physical en- 
vironment and to people show marked distinctiveness. (4) Distinctive 
qualities noticed early in life tend to persist; the child seems predis- 
posed to learn certain modes of adjustment and to reject others. Even 
before these adaptive forms are clearly defined an observer can often 
by the method of “prophecy” predict later traits. Irrespective of the 
methods used to study the consistency of early development the evi- 
dence is positive in virtually every case. 

This last conclusion, however, should not be regarded as imply- 
ing that personality is fixed once and for all during the first year of 
life. No one’s destiny is determined so early. Later circumstances 
affect personality profoundly. Ill health, a marked change in home 
conditions, or traumatic experiences, may actually reverse the course 
of development indicated during the first year of life. Furthermore, 
in early infancy there are certain conditions that prevent the forma- 
tion of dispositions as stable as those to be formed later. There is, for 
example, a low degree of retentivity for conscious experiences, in fact 
such a low level of consciousness itself that the child has no distinct 
self-awareness. Consciousness of self later plays a primary role in 
organizing personality. Then too, though the infant learns rapidly, 
he also forgets rapidly. Habits, the balance wheel of personality, are 
readily lost. Furthermore, his emotional life is not graded, being 
almost incapable of moderate response. The propensity for an all-or- 
none type of emotional activity prevents his learning discriminative 
affective responses or the development of a hierarchy of likes and dis- 
likes. His capacity for conceptualization is slight. There are, then, 
plenty of reasons why personality should be less stable, less predict- 
able, and less consistent in the early months of life than at any later 
time. It is indeed never again as unorganized and unstructured as in 
infancy. But since even in the first year of life it is neither com- 
pletely unorganized nor unstable, one must expect the maturing per- 
sonality to show even greater self-consistency . 40 

40 This chapter endeavors to interpret evidence drawn from modern investiga- 
tions, in terms of its theoretical and systematic significance for the development of 
personality. Readers desiring a broader background of fact are referred to an excel- 
lent summary of research (1920-1935): M. C. Jones and B. S. Burks, “Personality 
Development in Childhood ” Monog. of the Soc. for Res . in Ch. Devel 1936, Vol. !< 
No. 4. 



Das Geeinte zu entzweien, das Entzweite zu einigen, ist das 
Leben der Natur. 


It is easy enough to mark off arbitrarily stages in the progressive 
development of personality from infancy through senility. One 
could distinguish the periods of babyhood, childhood, adolescence, 
maturity, and senescence, or make finer subdivisions according to 
choice. But whether there are seven ages of man as Shakespeare held, 
or three as the Sphinx would have it, or five, or nine, or seventy, 
is a rhetorical and not a psychological question. For the single person 
there is only one consecutive, uninterrupted course of life. 

It is easy enough, too, to commit another psychological abstrac- 
tion and speak of the separate “mechanisms of growth” whereby per- 
sonality advances from one stage to another, for example, “condition- 
ing,” “the law of exercise,” or “the law of effect.” But here too the 
homogeneity of the process of growth is misrepresented. As Goethe 
points out nature continually unites and divides, builds up and tears 
down, organizes and disorganizes, in such a complex way that no 
single mechanism can be considered to operate independently. 

But if instead of speaking of separate mechanisms of growth as if 
they were so many dei ex machina, the psychologist distinguishes 
merely aspects of the total process, his analysis will be better justified 
and less subject to misrepresentation. It is only with aspects that the 
psychologist can deal. With the total, homogeneous phenomenon of 
growth he has nothing to do. For at bottom this phenomenon is 
simply an unexplained property of all living substances, a problem 
for the philosopher not the psychologist. An aspective rather than a 
substantive analysis of the process, then, is the safest one to follow. 

In planning such an aspective analysis, however, there is no 
authoritative guide. The list of aspects here distinguished will serve 

1 3 2 


as well as any. Their adequacy will be seen as they are followed 
through in detail in this and in subsequent chapters. 

A List of Various Aspects of Growth Significant in the 
Development of Personality 








Inferiority and Compensation 
Psychoanalytic “mechanisms” 

Functional Autonomy 
Sudden Reorientation: trauma 
Extension of Self 

Self-objectification: insight and humor 
The personal W eltanschauung 


In the previous chapter it was pointed out that much of the 
infant’s initial behavior is random and diffuse, mere “mass action.” 
From this matrix in some way specialized skills and precise adaptive 
responses gradually become differentiated. A crude prototype is the 
case of the amoeba. This animal has no organs, no differentiated 
parts. By changing its entire shape it produces a mouth, a hand, a 
foot, or an arm, whenever they are needed. Its reactions are total. 
Higher in the phylogenetic scale the various biological functions 
sooner or later become specialized and assigned to more permanent 
segregated systems for which specific structures develop. These struc- 
tures execute in a more adaptive way the responses orig inall y made 
by the clumsy and inefficient pre-structuralized whole . 1 

Not only do structures become progressively specialized through 
the development of limbs and executive organs, but so too do nervous 
functions and all the adaptive patterns of behavior. From the massive 

x The evolution of such special structures in the life-history of an organism, 
and their growth and self-repair under the domination of integral functional needs, 
have been described by G. E. Coghill, Anatomy of Behavior , 1929. 


i33 . 

retraction or outreaching innervations supplied by the undeveloped 
brain of the infant, and from the random twisting and squirming pro- 
duced by the dissipation of nervous impulses, the child gradually 
attains greater refinement in movement. Vocal habits, pleading, coax- 
ing, facial expression, gesture, reaching, and later, working, spending, 
saving, collecting, repairing, commanding, dissimulating, combating, 
adapting socially— become his differentiated and more efficient patterns 
of behavior, proceeding from, and substituting themselves for, the 
original, inaccurate, total responses of his whole body. 


Figure 14 

The Differentiation of Functional Systems 

A psychological account of differentiation is given by Lewin. 2 
“The child to a greater extent than the adult is a dynamical unity. 
The infant acts first with its whole body and only gradually acquires 
the ability to execute part actions.” Unlike adults, Lewin goes on to 
say, the young child has only slight functional firmness in the boun- 
daries of the various systems. In the adult, movements, mannerisms, 
attitudes and traits are more definitive, more fixed, less confused with 
unrelated tendencies. This course of development may be represented 
diagrammatically in the manner of Lewin (Figure 14). The bounda- 
ries between the functional systems are weak in infancy, causing the 
child to react as a whole; likewise the barrier between the child and his 
environment is less firm, leaving him a prey to all manner of environ- 
mental stimuli that later in life will be inhibited. The weakness of this 
barrier likewise prevents the development of sharp self-consciousness 
in the first year or two of life. 

Motor tensions in a child are far more totalized and imperative 
than in a self -controlled (well-differentiated) adult. A child of three, 
for example, who is “set” to respond in some particular way cannot 

= K. Lewin, “Environmental Forces in Child Behavior and Development,” Hand- 
book of Child Psychology, 1st ed., 1931, chap. iv. 



delay his response even until the starting signal is given . 3 And every 
parent knows the exasperating insistence of a child who must wait 
twenty minutes to be read to, or who expects to leave for the circus 
in an hour. There is little capacity for delay. There is likewise little 
capacity for graded response. When the child acts he uses much 
more of his body than does the adult. When he is pleased he jumps 
up and down; when he is angry he is “mad all over.” His emotional 
behavior is not differentiated, and nearly all of his activities are ac- 
companied by “synkinesis” (auxiliary but meaningless movement ). 4 
Especially when he reads, writes, talks, or practices on the piano, he 
wriggles or fusses. Precise co-ordination and patient skill are beyond 
him. The whole development of neurodynamics, writes Luria, con- 
sists in the creation of a “functional barrier” reducing the excitation 
from the total nervous system to certain limited systems that enable 
the child to mobilize accurately the amount of energy and the precise 
reaction patterns required by the situation . 5 

Differentiation, involving as it does the selection of precise adap- 
tive movements in place of gross and indecisive activity of the whole 
body, obviously depends upon the operation of inhibition. The inhibi- 
tion of antagonistic or useless movements is the first step in acquir- 
ing adaptive skill. A certain amount of inhibition is automatic, since 
the effective innervation of one muscular system will perforce in- 
hibit an antagonistic system. It is therefore not strictly true to say 
that the infant’s activity is ever completely undifferentiated. Inhibi- 
tion occurs in the earliest movements of the limbs and the first ex- 
pression of emotion. The child cannot thrust his leg forward and 
retract it at the same time, nor can he smile and cry. Curious effects 
are seen when the two innervations are struggling for the final com- 
mon path, but ultimately one or the other will prevail or else the 
child must remain in a tense state of suspended animation. 

Ultimately, thanks to the refinements of inhibition, not only 
actions executed by antagonistic muscle systems come to suppress one 
another, but also auxiliary, wasteful, and irrelevant actions become 
eliminated. There is growing precision; each act increases in skill and 
in adaptive value even while it decreases in area, spread, and in the 

3 A. R. Luria, The Nature of Human Conflicts, trans. 1932, chap, x. 

4 Cf. G. W. AHport and P. E. Vernon, Studies in Expressive Movement, 1933, 
pp. 16-21* 

5 A, R. Luria, op. cit p. 342, The doctrine of differentiation is in essence a 
modern statement of the Spencerian principle that the process of life consists in 
breaking down diffuse homogeneity into coherent heterogeneity . 



consumption of energy. With the aid of selective inhibition the ran- 
dom vocalizations of the infant become reduced in range and variety 
until there remain only the narrower patterns of vocal response 
serviceable for communication and symbolic thought. Gross and vio- 
lent emotional expression becomes reduced through inhibition until in 
the adult nothing overt but a subtle gesture, a lifted brow, or a well- 
chosen word may remain. Inhibition destroys the amorphous and mo- 
notonous character of early behavior, and generates greater diversity 
in activity and greater firmness in the boundaries between activities, 
The chief lines of support for the doctrine of differentiation (or 
individuation , as it is sometimes called) have not yet been enumerated. 
Working with human fetuses secured through abortions, Minkowski 
discovered that only in the older fetuses were there any signs of seg- 
mental reflex action, and that even in these there was a great pre- 
dominance of slow, asymmetrical and arrhythmical mass activity. 
When stimulated cutaneously these organisms seemed to respond 
with their entire muscular equipment . 0 Coghill’s work with the 
Amblystoma is well known. He has shown that the specific reflex 
activity of the forelimbs, legs and gills occurs after, and as a deriva- 
tive of, diffuse bodily movement. In the embryo of the Amblystoma , 
for example, the predominant type of activity is an S-like swimming 
reaction. This method of locomotion puts stress upon certain segments 
of the body, and by so doing seems to cause nerves to invade those 
regions, while this invasion in turn seems to determine the location 
of the limbs. The local limb reflexes thus derive from the more uni- 
fied preceding reaction of the total organism. To state the theory 
in his own terms: 

Organ systems must be secondarily attained by a process of indi- 
viduation within a totally integrated system, and the development of 
the behavior patterns must be effected, not by an integration of 
independent reflexes, as is usually considered, but by a process of 
individuation within a total organismic system which is from the be- 
ginning of reaction integrated as a whole. The basic principle, there- 
fore, in the development of the nervous system of vertebrates appears 
to be the maintenance of the integrity of the individual while inde- 
pendencies are grooving up within it, and are, so to speak, struggling 

8 An accessible account of this work is given by G. E. Coghill, /. Gen. Psychol , 
1930, 3, 431-435, The literature on the field as a whole is summarized bv L. Car- 
michael, Handbook of Child Psychology, rev. ed.. 1933, chap. L 


for ascendancy among themselves and for dominance over the indi- 
vidual . 7 

Irwin studied the movements of infants during the first ten days 
of life in a situation enabling him to keep constant the external stimu- 
lating conditions. 8 The records were taken continuously, 24 hours a 
day, for four infants. It appears that segmental responses are less con- 
spicuous than mass action, and that any segmental response becom- 
ing intense tends to irradiate until it involves all visible organs of re- 
sponse, so that “literally everything seems to be going at once.” It is 
an inescapable conclusion that since mass activity is the prevailing 
mode of response in early life, learning must proceed at least in large 
part, as specialization or individuation of this crude activity. The 
author also contends that an organismic theory of development 
through differentiation is far preferable to the older theory of sum- 
mated, compounded, chained or integrated specific reflexes. 9 

The physiological process by which habits and dispositions be- 
come differentiated within a loose primitive field of response is by 
no means clear. Why should useless movements be inhibited? Why, 
for example, does the infant at first reach with arms, feet, and head 
for an object; then with arms alone, and finally with one arm? The 
problem is to account for the fact that afferent impulses gradually 
cease to spread to any excepting certain preferential paths. Smooth 
and effective movement gradually obliterates diffuse and random 
movement, except for occasional synkinesis, or “nervous” movement, 
persisting in the form of personal mannerisms. Undoubtedly the proc- 
ess is enormously complex. Some authors have stressed the successive 
myelinization of motor and sensory nerve tracts and nuclei, or other 
temporal differences in the maturation of motor and sensory seg- 
ments. 10 Some speak of differential rates of metabolism favoring first 

7 G. E. Coghill, “The Growth of Functional Neurones and Its Relation to the 
Development of Behavior,” Proc. Amer. Phil. Soc., 192 6, 65, 51-55. 

Coghill has made specific application of his theory to the structure of person- 
ality: “The higher and more complicated expression of this conflict concerns the 
integrity of personality. Here it is chiefly the cerebral cortex that is the seat of 
conflict between the partial and total patterns of integration. The various components 
of the personality, according to this hypothesis, just as local spinal reflexes, develop 
by individuation within the mechanism of total integration, and their normal func- 
tion depends upon their subordination to that mechanism,” J. Genet. Psychol., 1936, 
48, p. 19. 

8 O. C. Irwin, Genet. Psychol. Monog., 1930, 8, No. 1. 

9 O. C. Irwin, Psychol. Rev., 1932, 39, 189-201. 

10 F. Tilney and L. Casamajor, Arch. Neurol. & Psychiat., 1924, 12, 1-66. O. G 
twin, Psychol. Rev., 1932, 39, pp. 189-201. 



one locality and then another, until specialized and regional response 
has supplanted diffuseness . 11 Other behavioristically inclined writers 
emphasize the different intensities and different periods in which 
certain environmental stimuli become effective , 12 or call upon “neu- 
robiotaxis,” the “reflex circle,” and “algebraic summation” of com- 
peting and antagonistic motor impulses . 13 It is safest to assume that 
all the resources of physiology will be required to give a complete 
account of this intricate process. In addition the psychological law of 
effect according to which successful adaptive movements are retained 
and unsuccessful ones lost will be called upon. 

In the meantime, the significance of the doctrine of differentiation 
for personality must not be lost sight of. The concept of the reflex- 
arc, which it opposes, arose as a consequence of the cell theory in 
biology. Just as the cell was accepted as the structural unit, so the 
functional co-ordination of sensory, central and motor neuronic cells 
into a simple arc was regarded as the unit of behavior. The cell theory 
held the organism to be a sum-total of intricately connected cells. 
When this notion was abandoned in favor of a more systemic theory, 
the reflex theory of nervous activity fell into disrepute. Today, there- 
fore, differentiation, in spite of its vagueness, is a principle usually pre- 
ferred by genetic psychologists to the specious simplicity of integra- 
tion. Even behaviorists have swung into line. “Simple reflexes,” write 
Lashley and Ball, “elaborated by a combination in chain reflex arcs 
have proved of little value for an understanding of the more intricate 
problems in psychology.” “ 

Some writers fear that enthusiasm for the doctrine of mass action 
as the original form of behavior, and the belief that from it are de- 
rived more specific, segmental reflexes and habits, may go too far . 15 
A controversy is on foot. Although the conditioning and compound- 
ing of simple reflexes seems to be an inadequate theory of growth, 
that there is still such a cementing or integrating process cannot be 
denied. Dissociation as a law of development cannot entirely super- 
sede the more ancient law of association. 

This dispute is of special significance for the problem of con- 
sistency in personality. If, as implied in the diagram of Lewin’s theory 

11 C. M. Child, Individuality in Organisms, 1915; and Physiological Foundations 
of Behavior , 1924, 

12 Z. Y. Kuo, Psychol Rev., 1932, 39, 499-5 1 5* 

is £. B. Holt, Animal Drive and the Learning Process , 1931, 

1* K. S. Lashley and J. Ball, /. Comp . Psychol , 1929, 9, 7-107. 

is Cf. I. Pavlov, “The Reply of a Physiologist to Psychologists,” Psychol Rev, 



(Figure 14), personality comes in time to have finer, more differen- 
tiated, and firmer divisions, would it not follow that personality is 
never again as unified as it is in early infancy? Would it not be ex- 
pected that each individual grows more and more specific in his con- 
duct and less and less of a consistent unit? Differentiation, as the sole 
mechanism, would produce a kind of entropy or dissipation of per- 
sonality. That movement, language and emotional expression become 
more subtle and precise no one can deny; and this fact is of exceed- 
ingly great importance for the development of personality. But these 
specializations once acquired seem to influence one another and to 
join together into a closely-knit and expanding organization. The 
course of nature, as Goethe said, is not only to divide what is united, 
but also to unite what is divided. Integration is fully as important a 
principle as differentiation. 


Integration, like differentiation, is sometimes considered to be the 
supreme principle of growth. Both genetic psychology and mental 
hygiene make wide use of it, and in the latter field especially it has 
been exalted to the place of the most solemn Om. Whatever condi- 
tion makes for mental health is called “integrative,” whatever con- 
dition makes for mental difficulty is called “disintegrative.” The dev- 
otees, of course, never doubt that integration is an unalloyed good, 
and disintegration an incarnation of evil. 10 But it is not always clear 
to what psychophysical process all this incantation refers. 

The original significance of integration is best understood by re- 
ferring to the cell theory in biology. The initial fact is that a human 
body contains about ten trillion cells, over nine billion of which are 
found in the cortex. Somehow out of this bewildering array of ele- 
ments a relatively unified and stable personal life is constructed. The 
single cells cohere in such a way as to lose their independence of 
function. From the many there emerges the one; the motto implicit 
in integration is e phiribus umim. 

Even though a person’s life exhibits contradictory trends, even 
though the unity is never complete and final, it is nevertheless obvi- 
ous that the number of totally independent qualities is not very great. 

16 Cf. W. H. Burnham, The Normal Mind, 1924, and The Wholesome Per- 
sonality, 1932. In spite of their over-emphasis these books are to be recommended 
for the profusion of concrete applications of the principle of integration to the 
development of personality. 



Probably only a very few specific segmental reflexes remain unasso- 
ciated with the complex activities of that great integrative organ, the 
cortex. Within this organ the links and combinations are of such pro- 
fusion that every function seems joined in some way and to some 
degree with almost every other function. 

Just as those who choose to regard the behavior of the newborn 
baby chiefly as mass action, prefer as a consequence to endorse the 
doctrine of differentiation, so do those who see the infant’s equip- 
ment at birth consisting chiefly in specific reflexes, prefer the doc- 
trine of integration . 17 

The doctrine of integration easily lends itself to a conception of 
hierarchy in personality. The simplest possible integration would be 
of two nerve cells functioning together as a simple reflex arc because 
of some synaptic affinity between them. Whether or not there are 
any reflexes involving as few as two neurones is not known. C. S. 
Sherrington, the physiologist who above all others is responsible for 
the concept of integration, regards this limiting case of integration 
as a “convenient though improbable abstraction.” Similarly, at the 
opposite extreme, where from a final perfect integration a completely 
unified personality is supposed to emerge, one meets another con- 
venient though improbable abstraction. But between these ideal limit- 
ing cases there is ample room for the operation of actual integration. 
In ascending order of complexity one might distinguish a hierarchy 
of the levels produced by integration as follows: 

Conditioned reflexes, the simplest learned forms of adaptive be- 
havior involving substitution of associated stimuli for congenitally 
effective stimuli, with the result that the individual performs innate 
acts to altered stimulus situations. 

Habits, integrated systems of conditioned responses, involving al- 
tered responses as well as an extended range of effective conditioning, 
leading to fairly stereotyped forms of response in the face of recurrent 
situations of a similar type. 

Traits, more dynamic and flexible dispositions, resulting, at least 
in part, from the integration of specific habits, expressing character- 

17 V. M. Bechterev ( General Principles of Human Reflexology, trans. 1932), 
and J. B. Watson ( Psychology from the Standpoint of a Behaviorist, 1919), are two 
writers who regard personality above all else as an integration of separate reflex 
arcs. Bechterev holds that the combining of reflexes is the only guide needed, and 
Watson speaks of the reflex level of functioning as occurring first in infancy, fol- 
lowed, through virtue of integration, by the conditioned reflex level and by the 
habit level. Personality, for Watson, is synonymous with the integration of an in- 
dividual’s manual, visceral and laryngeal habits. 



istic modes of adaptation to one’s surroundings. Belonging to this level 
are the dispositions variously called sentiments, attitudes, values, com- 
plexes, and interests. 

Selves, systems of traits that are coherent among themselves, but 
are likely to vary in different situations. (Cf. the statement of James 
that a man “has as many different social selves as there are distinct 
groups of persons about whose opinion he cares.”) 

Personality, the progressive final integration of all the systems of 
response that represent an individual’s characteristic adjustments to his 
various environments. (Considered as a perfect integration this level 
represents the ideal final stage, never actually attained.) 

A relatively simple case may be taken as an illustration of the way 
the mechanism of integration is supposed to operate. The newborn 
infant has the reflex capacity of withdrawing the arm when the hand 
is brought into contact with a dangerously hot object. In creeping 
around the room, perhaps at the age of 10 months, he reaches out 
and touches a hot radiator, quickly withdrawing the arm to avoid the 
painful contact. Next time he sees the radiator he withdraws “in ad- 
vance,” inhibiting the impulse to touch its shiny surface. This is a 
simple instance of conditioning. 18 At the same time it is an elementary 
case of integration: the sight of the radiator, originally unconnected 
with the withdrawing tendency, is now effectively integrated with it, 
with the result that one somewhat elaborated functional system is 
created. What is more important, the burned child usually avoids the 
hot radiator ever after, without periodic rebuilding or reinforcing of 
the integration. Once established, the system of avoidance continues 
to operate as a dynamic whole. This functional autonomy of systems 
acquired through integration is often found, and turns out to be one 
of the most important principles in the psychology of personality 
(cf. Chapter VII). 

Usually integration is portrayed in quasi-physiological terms-a 
perilous proceeding. That there are physiological correlates of inte- 
gration involving complex spatial, temporal, histological, and electro- 
chemical factors no one will deny. But since integration implies func- 
tional joining of nervous pathways, taking place presumably in the 
region of the synapse, and since this entire process lies still in the 

18 The law of the conditioned reflex may be stated as follows: Whenever a 
stimulus has a motor outlet, any stimulus occurring simultaneously will tend to ac- 
quire the same motor outlet; after sufficient repetition (sometimes one occasion is 
enough) the second stimulus alone will suffice to produce a discharge in that motor 



limbo of scientific mystery, all accounts of integrative growth in 
physiological terms are at the present time highly speculative. The 
characteristics of integration are known far better through psycho- 
logical investigations. 19 

To escape from the deceptive snares of pseudophysiology it 
will be safer to take from now on a frankly psychological ap- 
proach to the whole course of growth, and here specifically to the 

Figure 15 

A Schematic Representation of Integration 

problem of integration. In so doing a diagram will be helpful in sum- 
marizing the various phenomena that psychologists have classed as 
instances of integration (or of disintegration). Remembering always 
that integration is only one aspect of growth, and not the whole 
story, it becomes possible nevertheless to draw from it much help 
in understanding the development of personality. The significant 
features of Figure 15 may be analyzed under ten headings. 

19 Indeed, for that matter, the study of all the aspects of the development of 
personality may be said to be more advanced on the psychological level than on 
the physiological. “To reject the resource of psychological analysis and construct 
the theory of the mind solely on such data as physiology at present affords, seems 
to me a great error in principle, and an even more serious one in practice. Imperfect 
as is the science of mind, I do not scruple to affirm that it is in a considerably 
more advanced state than that portion of physiology which corresponds to it; and 
to discard the former for the latter appears to me an infringement of the true 
canons of inductive philosophy. . * ” J. S. Mill, System of Logic, Bk. VI, chap. iv. 
Sec, 2. This judgment made one hundred years ago is as appropriate today as when 
it was first written. 


(i) The Hierarchical Organization of Personality . Integration 
means that from disparate units of behavior larger and more Inclusive 
integers are formed. The actual functional scope of these new integers 
may be narrow or broad. A conditioned reflex represents an integra- 
tion of two or more sensory pathways (or types of excitation) with 
one common motor outlet; at the other extreme, personality itself 
represents an hypothetical integer including in a functional unity all 
the varied systems of behavior possessed by one individual For con- 
venience Intermediate levels may be distinguished, e.g., habits , traits, 
and selves , representing progressively widening integers of generalized 
dispositions. Actually, of course, these levels are arbitrary since inclu- 
siveness of any degree may characterize an Integration. 

The inborn reflexes, before and after conditioning, represent the 
simplest form of adjustive response. With time most of these become 
intricately joined into systems of habits, which though not entirely 
independent, have specific reference to one class of stimulus situations. 
But habits are still too rigid and too specific to serve as the most typical 
structural units in personality. Habits become functionally grouped 
and saturated with common characters. The resulting, and most Im- 
portant of all levels in the structure of personality, is the trait-level 
It is sometimes possible to speak further of organization in terms of 
“selves,” for sometimes, though seldom, whole systems of traits (each 
system constituting a “self”) are evoked separately under different 
environmental or psychological conditions. Finally, there is the theo- 
retical possibility of a thoroughly unified system of personality at the 
top of the pyramid. Whether this theoretical possibility is ever at- 
tained will be discussed later. 

(2) The Chronological Character of Integration . The base-line of 
the figure is roughly chronological It represents the process of sep- 
arate adjustive acts that the Individual is compelled to perform from 
birth to old age. An approximate division of this line into important 
“ages” of life is indicated. The pre-school environment is ordinarily 
constant and is responsible for calling out the initial and basic adjust- 
ments that form some of the habits enduring for a lifetime. Entering 
school introduces the child to many new and critical situations. Here 
he must adjust to individuals outside the family. As he approaches 
adolescence there is the “gang” and his own status among his contem- 
poraries. Adolescence itself brings a welter of physiological and social 
emergencies. Breaking away from the ties of home, perhaps leaving 
the shelter of home for good, precipitates a new crisis. A special sec- 
tion on the chart indicates the flood of intellectual, emotional and 
social experiences that follow an induction into the world of abstract 
thought, of industry, or that attend religious conversion. Such ex- 


periences, intangible though they are, have a profound effect on the 
maturing personality. 

As maturity advances there are other adjustments to be made, to 
one’s occupation, to one’s partner in marriage, and to one’s newly 
founded family. When this period is passed the personality is virtually 
finished. “If you have an adequate picture of the average individual at 
thirty you will have it with few changes for the rest of that indi- 
vidual’s life—as most lives are lived.” 20 By the age of thirty the “char- 
acter” has “set like plaster, and will never soften again.” 21 The verdict 
is perhaps too pessimistic and overdrawn, for some personalities seem 
to change markedly after the age of thirty. But in principle the judg- 
ment is sound, and is admitted to the diagram by assigning on the base 
line relatively small space after this age for the further integration of 
new qualities of personality. 

(3) Richness of Personality . The length of the base line is a 
measure of the range of experiences encountered in the course of life. 
It indicates the variety of adjustments the individual is forced to make. 
For some individuals the crises of life are lightly passed over; they 
bring no strain and leave no mark. In many cases, for example, the 
highly significant introduction to the abstract world of ideas never 
takes place. Some persons find it easier to keep an integrated view of 
life, because life has put few demands in their way. The country 
grandmother, lovable though she is, possesses only a few dominant 
habits and traits. She worries neither about the dictates of fashion nor 
the collapse of Capitalism; it is less important to her that the universe 
is wearing down than that her kitchen needs refurbishing. A few 
simple attitudes and rules of life serve her. She performs her daily 
duties, trusts in God, and drinks tea of herbs that she has gathered. 
Compared with an educated citizen of the world, buffeted about by 
discordant doctrines, tom by conflicts, personal and cosmic, her per- 
sonality is not many-sided and rich, though in all probability it is 
better integrated. 

(4) Incompleteness of Integration. By virtue of integration the 
rudimentary forms of adjustment in infancy commence to interlace, 
to form higher units with the passage of time. But the process is 
highly irregular. The weight of the various adjustments entering into 
higher units varies markedly. A bitter disappointment or grief, or a 
timely success may become the focal point for all future organization 
and thus serve to set a habit or a trait. On the other hand, many 
(perhaps most) of one’s experiences in life are never adequately 
integrated; they occur-a passing adjustment is made-but then 

20 J. B. Watson, Behaviorism, 1924, p. 22;. 

21 W. James, Principles of Psychology, 1890, 1 , p. m. 


the matter is dropped and forgotten. One's assimilated experience 
(Erlebms) does not keep pace with one's passing and transient ex- 
perience ( Erfahrung ). 

Bergson has said that personality is a knife edge pressed against the 
future. Constantly we encounter new events to which we must re- 
spond, but in responding we do not always incorporate the action 
into the permanent structure of our personalities. Rather we let our 
earlier habits and our previously formed attitudes and traits suffice, 
not troubling to alter our integrations to embrace the new features of 
the environment or the new evidence that we have encountered. Dia- 
grammatically expressed, there are innumerable x’s on the base line that 
never contribute to broadening a pre-existing integration or to the 
creation of one that is new and more adequate. 

This fact is of paramount importance. It is well known that men 
ordinarily direct their behavior by inappropriate habits, by stereotyped 
ideas, and by hollow verbal symbols. Most people do not learn as 
much from their own experience as they think. There seems to be an 
inertia in the process of integration. A few conventional habits, some 
fossils of ancestral political-economic beliefs, a handful of supersti- 
tions, a vocabulary of cliches, a simple W eltanschammg, serve most 
people to their satisfaction. 

The rule seems to be that unless some strong desire exists to alter 
an unsatisfactory habit or trait, or unless the demands of the world 
are so insistent that one cannot possibly continue to use his former 
equipment, or unless for some other reason an individual is genuinely 
pliable and open-minded, his personality will continue to employ its 
rough and ready devices for meeting the demands of life, thus avoid- 
ing the necessity for integration. 

(5) Regression and Dissociation. Two types of disintegration re- 
flect the difficulty of maintaining a progressingly unified set of atti- 
tudes and traits sensitive to the successive demands of life. Sometimes 
when an individual encounters harsh experiences, difficult or impos- 
sible to adjust to, his personality reverts or regresses to earlier levels 
of integration. This regression has both its normal and its abnormal 
manifestations. 22 It is not uncommon for a woman, finding it impos- 
sible to meet the demands of her married life, to return temporarily to 
the protective home of her parents; or for an adult defeated in some 
emotional encounter, to give way to tantrums or to coaxing; or for 
the Old Grad or the Legionnaire to seek to regain temporarily at a 
reunion the more carefree habits and outlook of his youth; or for an 

22 A many-sided treatment of the subject is that of F. L. Wells, “Social Mal- 
adjustments: Adaptive Regression,” in the Handbook of Social Psychology , 1935, 
chap, xviii. 


aged person partially childish, to resume the memories and even the 
habits of his early years. In abnormal instances usually after severe 
traumatic defeat, there is sometimes complete regression: a personality 
tumbles like a house of cards, leaving the victim as helpless as an in- 
fant. 23 

The process of dissociation occurs when some self-coherent system 
fails to integrate with the remainder of the personal life, becoming 
instead an independent “complex,” a logic-proof organization that re- 
sists the healing incursion of common sense and of other potentially 
neutralizing experiences. A phobia, an amnesia, an anxiety neurosis, 
are examples. Such separate systems indirectly affect the normal and 
primary systems controlling the personal life, causing severe conflict 
and giving rise to neurotic symptoms familiar enough in these post- 
Freudian days. 24 Such systems are often like mental cancers. In the 
diagram an attempt has been made to represent an independent, dis- 
sociated system of this sort in the minor integration marked (A). 

(6) Infantilism . Occasionally one meets someone who refuses to 
grow up, to “be his age.” Having found an earlier level of integration 
emotionally adequate, or perhaps for some reason fearing (probably 
unconsciously) to assume the burdens of adult estate, this person re- 
mains behind in development. This failure to grow is a counterpart of 
regression where maturity, once attained, is surrendered; in infantil- 
ism, maturity is not reached. 

A few years ago a freshman in college, the only son of a self-made 
man, was summoned before the dean for failure in his work. The dean 
asked if he was ’28, meaning the class of 1928. The boy replied, “Oh, 
no, sir, I’m only 18.” By the Freudian theory this might be regarded 
as a significant and ominous misunderstanding. Indeed, it developed 
that the boy who appeared like a fifteen year old, was attempting to 
win his way by an attractive smile, by coaxing, by childish appeals 
for sympathy, rather than by mental exertion and study. Consultation 
with the parents revealed that this mode of adjustment was always 
used by the boy at home, with marked success. In the diagram it is 
as though the personality had developed only to point (B) instead 
of to (B ') 9 the point suitable to the boy’s age. On the dean’s advice 
the lad was sent out to earn his own living and make his own way, 
with the result that in a few years the boy’s personality caught up to 
his years. This is a minor and not a spectacular case of infantilism. But 
it is of just such slight deviations that the variant patterns of normal 
personality are composed. 

as A dramatic and highly instructive case of this order is offered by W. Mc- 
Dougall, Case 24 in Abnormal Psychology , 1926, pp. 285-289. 

24 An excellent case of a dissociated phobic system is given in the autobiography 
of W. E. Leonard, The Locomotive God, 1927. 



(7) The Nature of Traits . Figure 15 also gives a hint concerning 
the nature of traits. It is apparent from the diagram that traits are 
more inclusive than habits, being compounded of them and of the 
underlying, still more specific, modes of adjustment. Traits appear 
gradually in life, becoming strengthened or refashioned as the base 
line of experience is extended. They are not wholly independent of 
one another. In fact, any specific instance of conduct may result from 
the confluent operation of several traits, and it may in turn contribute 
to the reorganization of these same traits. Traits are simply nodal 
points in the structure of personality, regions of adaptive stress, foci 
of adjustment (cf. Chapters XI and XII). 

(8) Multiple Selves . Different situations, as William James pointed 
out, and as sociologists constantly reiterate, may call into play differ- 
ent combinations and proportions of traits. At home a man may seem 
domineering, testy, and gruff; at work, considerate, tactful, even ob- 
sequious. His two major environments accent different complements 
of traits in one and the same personality. 

When the case is very extreme, when, like Dr. Jekyll and Mr. 
Hyde, the same individual has utterly incompatible and functionally 
separated integrations one may speak of dual personality . In rare cases 
more than two such independent systems have been clinically re- 
corded; these are multiple personalities. In Figure 15 the possibilities 
of “selves” turning into multiple personalities is indicated by the 
pyramids (C and C')« 

In general, however, the situation is reversed. Even the man in the 
above illustration probably shows more characteristics that are alike 
in his two environments than are unlike. The case for the existence of 
separate and distinct selves is easy to exaggerate. James is partially re- 
sponsible for this exaggeration because of the quotable aphorism, a 
man “has as many different social selves as there are distinct groups 
of persons about whose opinion he cares,” 25 which decidedly over- 
states the case. 

(9) Rigidity versus Flexibility . Is the well-integrated person able to 
vary his behavior as occasion demands, or is his character really set 
like plaster? Some writers regard integration as synonymous with 
petrification: “With the perfection of reaction patterns achieved dur- 
ing advancing years, personality becomes better integrated, but a 
heavy price is paid: imagination fails and intellectual sterility en- 
sues.” 20 

It is true that by its very nature integration reduces the irregular 
25 'Principles of Psychology, 1890, I, p. 294. 

2S J. T. MacCurdy, Common Principles in Psychology arid Physiology , 1928. 


and random nature of adjustment by providing definite guiding sys- 
tems within which responses are prepared. The drifter, the soldier of 
fortune, the person who readily takes on the complexion of his en- 
vironment, cannot be said to have a firmly integrated personality. No 
one, as William James remarked, can at one time be a bon vivant, a 
philosopher, a tone poet, a lady-killer, and a saint; such diverse inte- 
grations could not keep house together in the same tenement of clay. 
Even if these phases succeeded one another in the course of life die 
picture would still be one of pathological disunity. 

Yet integration does not necessarily mean rigidity in personality. 
Flexibility may exist side by side with integration. It all depends upon 
the nature of the integration itself. If immature, rutted, intolerant sys- 
tems have developed and become set, they will not readily bend to 
new experiences nor be modified by them. But if a system is held 
somewhat tentatively, or if it is broad and ample in scope, the inte- 
gration itself will not preclude adaptability in conduct and progressive 
change in personality. 

(io) The Unity of Personality . Rarely does one encounter a per- 
son who seems to be completely integrated with only one dominant 
philosophy of life from which every attitude, trait, and individual act 
must proceed, and for whom perfect prediction of conduct is 
possible. Occasionally a literary writer argues for such a case. 27 And 
there are some psychologists who contend that if a personality is 
rightly understood It will always be found to lie under the domination 
of one controlling goal, one ruling passion, possessing (to borrow 
Professor Wertheimer’s term) one radix. To allow for the possibility 
of such a supreme, all-inclusive integration, the pyramid in Figure 15 
is closed with a dotted line (D). But the problem of the unity of per- 
sonality is not to be disposed of so easily. In Chapter XIII we shall 
return to it. 


Maturation may be defined as the ripening of innate behaviorial 
tendencies in the absence of training and experience. The ripening 
may involve a native process of differentiation , of integration , or of 
both. Since the visible equipment of the infant at birth is relatively 
meager (Figure 1 1 ), supplying only a partial demonstration of the 
child’s hereditary endowment, the conception of maturation has of 

£7 For example, a statement from a London newspaper: “Einstein is all of a 
piece. ... His social behavior, his political views, his general interests, all seep to 
be the spontaneous manifestations of a profoundly integrated nature. There is no 
division in him.” Another instance is Chesterton’s keen characterization of Tolstoy, 
cited on pp. 190 f. 


necessity been embraced by those who are interested in the native 
factors in growth. It is obvious that the more one leans on heredity 
the more must one depend on maturation. Instinct-psychologists are 
an instance in point; since the newborn infant shows no visible signs 
of a parental, acquisitive, or gregarious propensity, they have to rest 
their case for these instincts upon the latent powers of maturation. 
Opponents of this doctrine reply that subtle and socialized types of 
learning account for the appearance of these “instincts” later in life . 28 
The subject requires further consideration. 

At the outset it must be granted that there is indisputable evidence 
in favor of two kinds of maturation, though neither of these has any 
very specific relation to the traits of personality. In the first place, it 
is clear that the nervous system as a whole ripens in its capacity for 
the retention of experience and for intelligent adaptation to new situ- 
ations. Intelligence is certainly not complete at birth. Normally it in- 
creases in “power” through adolescence, probably later. Likewise, 
certain intellectual capacities— talents for art, music, mathematics, and 
the li k e— we must admit as inborn and as undergoing maturation. 
This type of growth-from-within parallels, approximately, physical 
growth. There is also an obvious ripening of glandular functions, 
most notably those regulating sexual development. Granting all this 
physical, intellectual, and sexual ripening, there are still no proofs 
of the unfolding of innately determined instincts nor of pre-ordained 
traits of personality. However important the intellectual capacity 
and sexual drive may be in the development of traits, still no adaptive 
trait is ever given in advance. The ripening sexual impulses, for ex- 
ample, will be met within the framework of pre-existing attitudes 
and fears, and will be affected by traits acquired well before the onset 
of puberty. 

Another type of maturation indisputably occurs— an object of 
much investigation in recent years— the unlearned ripening of motor 
co-ordinations and capacities such as creeping, walking, climbing, 
swimming, laughter, and vocalization. The ingenious development of 
the method of co-twin control has contributed to this research. Iden- 
tical twins (with presumably equivalent heredity) are separated for 
the purposes of experiment. One receives an intensive course of train- 

28 A statement of the ease for maturation from the standpoint of instinct-psy- 
chology is given by W. McDougall, Outline of Psychology, 1923, chaps, iv and v. 
The case in opposition is stated by F. H. Allport, Social Psychology, 1924, pp. 44-48. 
A criticism of the opposition is offered by R. S. Woodworth, J. Abnornt. & Soc, 
Psychol., 1925, 20, pp. 94-98. 


1 49 

ing during the early months of infancy, say in walking or in climb- 
ing; the other remains untrained, or perhaps is, so far as possible, pre- 
vented from learning. The results commonly show that at a given age 
both infants will perform these chosen functions almost equally well; 
it was growth from within, not outer training that developed the 
ability . 29 It would seem that until certain nerve tracts myelinize, or 
certain nerve centers mature, these activities are impossible, but that 
once the inner processes of growth have prepared the appropriate 
nervous mechanisms these motor functions appear regardless of ex- 
perience or training. 

Besides the method of co-twin control, careful observation of 
babies supplies plenty of ground for the same conclusions. The evi- 
dence is of two kinds . 80 In the first place, all babies seem to go 
through the same stages of development (e.g., in locomotion and 
vocalization) regardless of efforts to change the order of the appear- 
ance of these stages. The length of a given stage for some babies may 
be long, and for others short, but the order is the same. In the second 
place, the development of such functions appear with a suddenness 
not at all common in motor learning. This suddenness is unlike the 
occurrence of “insightful” behavior, for it occurs when there is a 
minimum development of appropriate anatomical structure, and when 
there has been a complete absence of training. 

If, in line with available evidence, the process of maturation is 
regarded as limited to the general ripening of somatic and nervous 
structures, and to the ripening of a few rather specific locomotor and 
vocal functions, it cannot be regarded as one of the direct fashioners 
of personality. Gesell is in danger of overstating the case when he as- 
serts that “maturation must be granted a basic role even in the pat- 
terning of personality and career .” 31 A man’s personality and career 
are not due primarily to what ripens within him but to the manner 
in which he lays hold of these maturing functions and incorporates 
them into what he has already learned. 

It is sometimes said of a grown man that he more and more re- 
sembles his father as he gets older. Tastes, mannerisms, and attitudes 

29 Illustrative experimental studies of this order are: M. B. McGrow, Growth; 
A Study of Johnny and Jimmy, 1935; L. C, Strayer, “Language and Growth: The 
Relation of Early and Deferred Vocabulary Training,” Genet. Psychol. Monog., 
1930, 8, No. 3; A. Gesell and H. Thompson, Infant Behavior, Its Genesis and Growth, 

80 M. Shirley, The First Two Tears, 3 Vols., 193W933. 

31 A. Gesell. The Guidance of Mental Growth in Infant and Child, 1930, p. 192. 



repeat the parental pattern, even though the direct parental influence 
has many years been removed. Might it not be that inherited traits 
lie dormant and rise to take form within the personality many years 
after childhood and after the period of general maturation have 

of course a possibility, but still there are two strong adverse 
arguments. (1) How does one know that subtle learning has not 
taken place, that ideals early implanted take their form when appro- 
priate situations arise? From her own training the daughter learns 
how to bring up children, but she cannot put this knowledge into 
effect until she has children of her own; nor can a son or daughter 
follow the parents’ example in household management, citizenship, 
or club membership until there is an opportunity to do so. (2) As the 
physique and its inherited deficiencies reach their mature expression, 
certain appropriate forms of adjustment must occur; but it is the 
physique and its deficiencies, and not the forms of adjustment that 
mature. An inclination to obesity or to baldness, may be a late conse- 
quence of heredity, and as one comes to look more like the parent, 
it is likely that similar physical demands, provided the environment 
too is similar, will lead one to act more like the parent. “Like father, 
like son,” is a proverb not merely of heredity and maturation, but also 
of example and learning. 

Maturation contributes to the development of personality by 
bringing out every inherited feature (pp. 107-109 and Figure 11). 
These features include physical structure, peculiarities of tempera- 
ment and talent, the general capacity for intelligent modification of 
behavior, peculiarities of physical growth and decay, the latent sexual 
functions, and numerous specific locomotor and vocal patterns. All 
these develop by virtue of an inherent maturational capacity. But none 
of these qualities are in themselves independent units of personality. 
They contribute to the formation of personal dispositions, but their 
influence must be combined with the demands made by the environ- 
ment upon the individual. Goals and purposes are not inherited, unless 
one grants that vague primordial need-to-live. Special interests and 
so-called “instincts” develop, as do traits and attitudes, through the 
many-sided effort of the individual to find a balanced position for 
himself in the world he has to live in. Maturation presents him with 
new internal situations to which he must adjust; but excepting at a 
rudimentary motor level it does not provide him with ready-made 
instruments for the task. 

It is 


I 5 I 


Taken broadly, the field of learning includes every form of ac- 
quisition and modification that occurs in the course of growth. Every 
way of learning is at the same time a way of building or of changing 
one’s traits, and hence, at first glance, it seems reasonable to equate 
the problem of the development of personality with the problem of 
learning. But traditionally in psychology what is called “learning” 
is a somewhat narrower problem; though even in its narrower sense it 
forms one of the longest and most disputed chapters in psychology. 
Since the account here must be kept within bounds, only three of the 
most pertinent applications of “learning” to personality will be re- 
viewed, viz., conditioning , efferent modification, and imitation. The 
three succeeding chapters, taking a broader view of learning, will 
amplify the account. 

Conditioning. At times this concept is employed in the loosest pos- 
sible way, as a pseudo-explanation of all learning. Properly used, how- 
ever, it refers only to the extension of the range of stimuli that will 
arouse a given response. The law of the conditioned reflex, stated on 
p. 140, deals only with the fact that a relatively simple response orig- 
inally initiated by a certain stimulus, A, is now initiated by another 
stimulus, B, that has occurred frequently in connection with A. It is 
important to note that the response is not altered by the conditioning, 
but is simply reinstated by a series of secondary (conditioned) stimuli. 
Conditioning is thus at most a theory of “afferent” learning; it ex- 
plains why a wider and wider range of stimuli are reacted to, but not 
how the reaction itself becomes modified, made more precise and 

Any theory of personality requires some principle to account for 
the widening of one’s tastes, interests, desires and aversions, in the 
course of growth. Conditioning serves the purpose, in part, although 
one may reasonably debate whether it is any great improvement upon 
its predecessor, the venerable doctrine of the association of ideas, or 
whether it will not eventually be replaced by some more discriminat- 
ing theory of mental organization. 

In discussing the biological theory of personality (pp. 1 14-12 1) the 
principle was applied to the rapidly widening range of stimuli that 
arouse drives. The primitive segmental tensions, the theory maintains, 
are easily conditioned, so that acquisitive behavior is aroused not only 


by a few prepotent stimuli to which the infant responds reflexly, but 
by a wide range of desirable objects in the environment. Conversely, 
the negative valences of the environment, the range of objects that are 
rejected and avoided, is extended through the same principle. 

A simple illustrative case might be the establishment of liking for 
a certain color. Suppose a man decorates his room in blue, is also un- 
usually fond of blue in clothes, and plants many blue flowers in 
his garden. This taste is an ingrained characteristic of his personality; 
he has an acquired disposition to perceive, to seek out, and to respond 


Development of a Taste Through Conditioning 

to this color in a favorable way. A hypothetical account of the case 
in terms of conditioning might lie in the association of this color 
through two or three connecting links to some original cause of 
favorable or adient behavior, as suggested in Figure x 6. Such a proc- 
ess of conditioning, or association through contiguity in time, might 
account also for a person’s liking of certain poems, churches, songs, 
pictures, vocations, philosophies of life, or traits of personality, or 
for his repugnance to certain foods, races of people, or philosophical 
and moral doctrines. 

That conditioning is not the whole of learning is easily seen from 
a glance at Figure 16, for certainly when he is mature this man does 
not respond in the same way as he did in infancy. He does not com- 
mence to mouth or suck or nestle when confronted by the blue fab- 
rics or blue flowers, although this type of response is all that the doc- 
trine of conditioning taken by itself would expect. Plainly the process 
of learning involves not only an extension of the range of effective 
stimuli , but also an alteration in the response; there is not only afferent 
learning, but efferent learning as well. 



There are several difficulties with the doctrine of conditioning 
even as a theory of afferent learning. For one thing, it implies that 
any stimuli occurring simultaneously with a pre-existing effective 
stimulus will acquire the same motor outlet. This implication is mani- 
festly false, for innumerable surrounding concurrent stimuli fail to 
act as conditioners. Strong emotion or deliberate effort on our part 
seems to enhance the chances of successful conditioning, but the mere 
temporal contiguity the law assumes is insufficient. Furthermore, the 
conditioned reflex as it is discovered in the laboratory— the only place 
where it is known thoroughly-is a capricious affair. It is easily extin- 
guished, dies if it is not reinforced by the original unconditioned 
stimulus at intervals. Competing conditioned stimuli not infrequently 
cause the experimental animal in laboratory situations to fall asleep. 
So far as we can observe none of these capricious phenomena exist in 
the conditioned responses of normal human life. 

All in all, it is best for the present to think of conditioning as a 
broad principle, crudely formulated, and too coarse for precise appli- 
cation in the field of personality. By itself it does not do justice to 
the many selective functions of the human mind that have the power 
of limiting and directing the course of conditioning. On the other 
hand, as a rough modem equivalent for the indispensable principle 
of association, it has its uses, and seems to express rather more dynami- 
cally than the older law the fact that stimuli associated with active 
systems often forge a functional bond with these systems. 

Efferent Modification. Two of the familiar and traditional laws 
of learning are frequency and recency. The former simply restates 
the adage “practice makes perfect,” and the latter calls attention to 
the obvious fact that we tend to remember recent events and newly 
acquired skills better than those of long ago. Though undoubtedly 
valid, these principles cannot account for change in behavior. By 
themselves they would produce complete stereotyping of conduct; an 
act recently performed would be repeated and the more it was re- 
peated the stronger would become the tendency to continue it! 

Within the traditional psychology of “learning” the improvement 
in adaptive behavior has been left to the law of effect, according to 
which successful modes of response are retained and unsuccessful 
modes are lost. Since success is usually accompanied by pleasure, and 
failure by pain, the law is often stated in a hedonistic fashion: pleasure 
stamps in, pain stamps out. Troland has cast this theory into a quasi- 
physiological form. Successful acts, he believes, characteristically re- 


suit in stimulation of “beneceptors ” giving to the organism pleasur- 
able sensations, and sending back to the cortex characteristic afferent 
excitations having the property of enhancing the conductivity of the 
motor pathways then in action. In this way successful acts are 
“stamped in.” Conversely, “nociceptors” are stimulated by acts that 
bring baleful consequences to the organism, or at least consequences 
immediately disagreeable. The return afferent impulses from these 
sense organs have the property of heightening the resistance at the 
motor centers, so that the same act is less likely to be repeated a 
second time . 32 

According to this hedonistic formulation, personality results from 
the retention of those modes of adjustment that have yielded pleasure, 
and from the habitual avoidance of events that bring displeasure. 
“Strong personalities are those which have been consistently built 
upon a coherent foundation, in which the pull of pleasure and the 
push of pain are in the same direction, when this is the direction 
which leads to success .” 38 Since for most personalities, the pull and 
push are not in the same direction, the individual can only obtain the 
best possible hedonistic balance, the greatest possible pleasure with 
the minimum of pain. In any case the operation of the law of effect 
is thought to be quite automatic. Responses yielding pleasure are 
retained; those resulting in pain are abandoned. Personality grows 
through the production of tentative trial-and-error acts, some of 
which are selected by the Great God Pleasure for establishment, and 
some by the Great God Pain for banishment. If this picture were 
true mortals should be more content with their personalities than they 
are, for all annoying and unsatisfactory habits and traits in oneself 
would be automatically eliminated. 

The chief fault of hedonism is that it confuses the by-product of 
a complex process with the process itself. Pleasure usually accom- 
panies successful striving, but it is not necessarily the goal of the 
striving. It is more often an indicator that a successful mode has been 
discovered. There is no evidence that this mode is automatically 
selected for preservation. The hedonists’ account of personality as a 
jellification of pleasure-giving habits is far too simple. 

A more liberal interpretation of the law of effect is offered by the 
doctrine of Insight, whereby human beings are credited with the 
ability to survey and organize the confused field that confronts them, 

82 L. T. Troland, Fundamentals of Human Motivation, 1928. 

38 L. T. Troland, Mystery of Mind, 192 6, p. 171. 


! S5 

intelligently comprehending the potential relationship of the various 
factors to one another, and finally adopting the course of behavior 
most suitable to their needs. By virtue of this ability one line of con- 
duct is understood to be more suitable than another and is accord- 
ingly adopted. (It must be assumed, of course, that what every indi- 
vidual is doing is to strive always for a more effective relationship 
between himself and his behavioral environment.) Even though in- 
sight may fail in its attempts to change the modes of adjustment or 
even though it may render them less suitable, still the very existence 
of insightful behavior presupposes that intelligent effort and evalua- 
tion underlie the phenomena of change in personality. 

This doctrine has the advantage of offsetting the merely mechani- 
cal laws of learning with a recognition of the oft-forgotten human 
capacities for foresight, planning, and problem solving. But by itself 
it gives a somewhat rationalistic view of the developing personality, 
and needs to be tempered with a recognition that not all the “clos- 
ures” formed in personality are intelligent structurations of the field. 
Many of them, as observed earlier, are cliches, complexes, and stereo- 
typed habits with which the individual bludgeons his way through a 
psychological environment too subtle for him to meet in a truly 
insightful way. 

Imitation. A child’s personality, and to a less extent an adult’s, is 
fashioned in part through imitation. That is to say, many modes of 
adjustment are taken over more or less ready-made from other per- 
sons who serve the imitator consciously or unconsciously as models 
for conduct. In the latter part of the nineteenth century this form of 
learning received so much emphasis that a simple and sovereign 
“instinct of imitation” was postulated as underlying the acquisition of 
all mental and motor dispositions. Imitation was conjured up to ex- 
plain everything (which was decidedly over-working the trick). 
Nowadays, instead of assuming one special power of imitation operat- 
ing relentlessly on all occasions it is better to distinguish at least three 
different processes, all imitative in their effects, but independent in 
their manner of operation. 

The first is a kind of conditioned reflex imitation, sometimes called 
the “echo principle.” 34 It is significant chiefly in the first year of life, 

84 E. B. Holt, Animal Drive and the Learning Process, 1931, states the principle 
as follows: “A child will learn to echo back any action of another, provided that 
another’s performance of the act stimulates any of the child’s sense-organs at a 
moment when the child is engaged in a (random) performance of the same act” 

(p. 112). r A-V vAA A y & : ; 


for to it may be ascribed the first linguistic forms that the child 
adopts as well as many early conventional gestures (of refusal, ac- 
quiescence, good-by, etc.). The stages are three: (a) an infant per- 
forms an act (e.g., a vocalization) quite “accidentally” (i.e., without 
social reference) ; (b) an adult reading into it some social significance, 
performs the same act at the same time in the child’s presence (in 
reality the adult imitates the child) ; (c) thereafter, by the principle 
of conditioning, the child may repeat the act when he hears or sees 
the adult performing it. Unquestionably, the first words that a child 
learns to speak are acquired in precisely these three stages . 35 Even 
later in life some simple forms of imitation seem explicable in the same 
way (e.g., laughing when another laughs, or yawning when another 

A second early form of imitation is less certainly established; it 
may even be reducible to the first, though it is rather unlikely. To 
illustrate, “an infant of a few months, lying in the mother’s arms, 
showed signs of restlessness and anxiety and wept silently when the 
mother became disturbed by a reference in the conversation.” Again, 
“an older child showed anxiety (shyness) when the mother was per- 
turbed by the presence of a stranger at the door. In such situations 
there seems to be some direct method for translation of attitudes, 
without the necessity of experiencing them. As a working premise, 
only, it may be assumed that, through vision or direct contact, a 
mimetic capacity in the individual is stimulated, and that he then 
takes into his own muscles the postures and tensions of the person ob- 
served or felt; and translates them in terms of muscular and glandular 
activity.” 88 This unconscious mimicry of “muscle tensions” in others 
is not well understood and needs further investigation. It has been 
assumed to play an important part in our understanding of other 
personalities (Chapter XIX) , and likewise to account for the growing 
resemblance between persons constantly associated with one another 
(e.g., husband and wife). 

The third form of imitation is the conscious and deliberate copy- 
ing of the behavior of another. It appears toward the close of the 
first year. A few instances from the psychological diary of little 
Andrew enlighten the point. 

85 F. H. Allport, Social Psychology, 1924, pp. 181-188. 

86 S. and M. Blanton, Child Guidance, 1927, p. 26. 



8.28 A. watches intently father smoking his pipe. When father offers 
A. the pipe (properly cleaned), A. puts the stem into his mouth 
and blows vigorously. 

According to the theory of conditioning, Andrew should have sucked 
the stem (as he would a nipple or pacifier). Actually Andrew imi- 
tated the father blowing out the smoke. The resemblance of his ac- 
tion to the puff of smoke he had been watching demands explanation 
in terms of insight— a deliberate attempt to reconstruct or to duplicate 
the stimulus-situation as understood. He could not, of course, know 
that smoking required sucking as well as blowing. This is the first 
of a long series of similar instances in this particular child’s history. 

14.8 A. always objects to having his nose wiped, but takes handker- 
chief out of his mother’s apron pocket, puts it to his own nose 
and snuffles. 

20.0 The doctor gave A. great discomfort by depressing his tongue 
with a spatula while examining his throat, causing A. to wriggle, 
cry, and reject the object. Fifteen minutes later, left to himself, 
A. picks up the spatula and puts it down his throat just as the 
doctor had done. 

In these instances Andrew is seen deliberately reinstating situations 
that originally caused annoyance or pain. Instead of rejecting the 
stimuli associated with this annoyance, he returns to examine them 
and to reinstate in part the annoying situation, quite contrary to the 
doctrine of conditioning and to the hedonistic law of effect. The de- 
sire to comprehend, to execute a meaningful act shows that learning 
through insightful imitation may transcend the mechanical principles 
alleged to account both for learning and for conditioned reflex imi- 
tation. 37 

Advancing to later childhood, deliberate copying becomes a more 
and more important factor in the growing personality, especially, for 
example, in the adoption of prejudices, beliefs, and attitudes of one’s 

A. (now nine years of age) heard many discussions of politics prior 
to a presidential election. Wondering about the merits of the various 
political parties, he came to his father and asked, “Daddy, what are 
we?” A-A-'U ■ ; V\ . ' 

sr It is not uncommon to find this insightful imitation combined with the un- 
conscious mimetic form of imitation described above. It often happens in personal 
relationships that each of two friends imitates the other more than he realizes. 



It seemed never to occur to the lad, who in much of his behavior is 
very independent, that he might decide the issue for himself. He had 
learned that the assumption of his parents’ attitudes was a fairly safe 
procedure when needing a guide. He begged to have his mind made 
up for him on a troublesome issue; he wished to imitate his parents’ 
partisanship. In so doing he is planting the seed for many later quali- 
ties in his personality. He may continue merely an uncritical “in- 
heritor” of his parents’ views, or, when childhood is past, he may 
rebel, adopting different views, perhaps after rebellion to return 
again to the point of view of his parents. In any case his entire life 
is affected by his conservative imitation of parents in his early years. 
The story can be duplicated in every child’s life, not only in the 
adoption of a political outlook, but likewise in religious affiliation, 
occupational and racial loyalties, moral and esthetic codes. 

This type of insightful imitation lasts throughout life. When an- 
other person seems to us to have worked out a happy solution of the 
problems that confront us, we often try to adopt in toto this solution, 
including sometimes the qualities of personality that make the solu- 
tion possible. Such imitation is a labor-saving device for us, a short 
cut in the development of our own personalities. Certainly it is sin- 
cere if unintentional flattery to the individual we imitate. 



Or the whole of our own natures we are never directly aware, nor 
of any large portion of the whole. At any single moment the range 
of consciousness is remarkably slight. It seems only a restless pencil 
point of light entirely insufficient to illuminate the edifice of person- 
ality. Yet, for all its feebleness, it provides each of us with the one 
and only sure criterion of our personal existence and identity. The 
past is drawn out in successive and overlapping conscious moments, 
backwards, twenty, thirty or forty years to early childhood, and the 
future extends, vaguely but still intimately, before us in each over- 
lapping moment of planning and imagination. It is through this dove- 
tailing of the successive moments of consciousness with their imbrica- 
tion of temporal reference and content, that we arrive at the convic- 
tion that we do somehow possess consistent personalities surrounding 
the momentary conscious core. Unless we postulated for ourselves 
a permanence of personality we could not possibly account for the 
many identical threads running through our conscious states . 1 


Consciousness and self-consciousness are not identical. Not all ex- 
perience possesses ego-reference . 2 Some of it simply occurs, the sub- 

i It is nowadays fashionable to distrust the evidence of immediate experience. 
Behaviorists, logical positivists and psychoanalysts have joined the discrediting 
chorus. Each individual, they say, is aware of so little, his introspection is so faulty 
and unverifiable, he is so given to self-deception, that direct awareness must not be 
admitted as a scientific datum of any importance. It is following this line of reason- 
ing that the Unconscious or the physico-chemical Bodily Constitution has been 
proposed as the true matrix of personality, the only region worth exploring. This 
reasoning has something to commend it, but even so, the core of the objective 
method is still the reliance each scientist places upon the testimony of his own fugi- 
tive and overlapping conscious states. He can work with the Unconscious or with 
Bodily Constitution only as they are distilled into his own consciousness. And what 
is more important, his acceptance and rejection of evidence, his devotion to his 
own standards, are bound to the still more subjective core of his personality, viz., 
his self-consciousness. 

2K. Koffka, Principles of Gestalt Psychology , 1935, p. 328. Koffka treats the 
Ego both as a phenomenal object (the direct object of knowledge) and as the phe- 


ject not feeling that the experience is somehow tied to his interests, 
to his memories, and to his personal life (though a slight shift of atti- 
tude can change almost any state of consciousness into a state of self- 
consciousness) . Genetically this fact is of considerable importance, 
for it is agreed that in the young child the consciousness of self is a 
gradual and difficult achievement, whereas consciousness of a less per- 
sonal order is no doubt present from birth. What the earliest con- 
sciousness, devoid of self-reference, is like, is difficult to imagine. 
Readers of William James are by now persuaded that it must be a 
“big, blooming, buzzing confusion”; though, as a matter of fact, it 
may not be so at all, but rather a consciousness of sharply segregated 
figures upon dim backgrounds as Gestalt theory would prefer to have 
us believe. In any case, the young infant seems quite unaware of him- 
self as a self. He treats his own body as if it were foreign to him; his 
toes are his toys, and he may claw his own face until it bleeds. He has 
no “bodily” self, no “social” self, and no “material” self. The bound- 
ary between him and not-him, between his and not-his is unestab- 
lished. There is, as Koffka would say, little or no Ego-organization. 

Until the child has a fairly definite conception of himself as an 
independent person he cannot conceptualize his relationship to the 
surrounding world, and hence lacks the subjective nucleus for the 
development of his own personality. By bringing the consciousness 
of the self only slowly into focus during the first three or four years 
of life nature seems to withhold from the individual the very key- 
stone to his structure of personality. 

Several conditions are responsible for the infant’s lack of self- 
consciousness. There is the well-known deficiency in his memory- 
life. Recognition, which occurs much earlier than recollection, is not 
clearly present until the latter part of the first year, and then it fails 
if the interval of time involved is more than three or four weeks. As 
for recollection, it is rare indeed to find accessible memories reaching 
back before the age of three, and never into the first year of life. 

nomenai ground against which many of our perceptions occur. The term “self- 
consciousness” has the same double reference, and may be considered essentially 
synonymous with Koffka’s Ego. But there are two reasons for preferring the former 
term, (i) Philosophy has previously equated Ego with the Knower, whereas Koffka, 
disregarding the nominative form of the Latin pronoun, equates it with the Known , 
with the empirical Me. (2) At the core of the Ego, thus conceived, Koffka places 
the Self, a sub-system where are located the deeper layers of the Ego, and strongly 
emotional states of personal reference (p. 342). This division does not seem to be 
necessary. The difference being entirely one of degree it can be ruled out in favor 
of one consistent account of the genesis and nature of the consciousness of the self , 
whatever its layers may be. 


Why are early memories so impermanent? One answer is that 
infantile experiences are not verbalized, and cannot therefore be held 
as concepts in consciousness. Another answer invokes maturation, 
holding that the areas of the cortex involved in conscious memory 
are undeveloped (specifically— unmyelinized) , and therefore unable 
to hold “traces” of experience. Motor areas, on the contrary, are in 
an advanced state of development during the early months of life, for 
it is well known that motor habits once learned are readily retained. 
Still another answer claims that recall depends upon the child’s ability 
to place an event in some context familiar to him, and that there is 
too little of this apperceptive context in early life to serve as a pre- 
servative of early experience. A fourth theory, the psychoanalytic, 
holds that the infant upon encountering the “reality principle” re- 
presses into the unconscious all memories of his self-indulgent life up 
to that date. However it comes about, this fugitive condition of mem- 
ory in early childhood is most certainly a handicap in the develop- 
ment of self-consciousness. 

Another handicap arises from the ungraded and undifferentiated 
character of the infant’s emotional responses. Affectively he behaves 
whole-heartedly or not at all. He does not evaluate the stimulus ac- 
cording to the degree of its importance in relation to his own needs 
and desires. He laughs and cries excessively, almost as an automaton. 
Such behavior can only be accounted for by assuming a virtual all-or- 
none operation in response to stimuli, devoid of the grading effects of 
inhibition that would intervene if each stimulus were tested out by its 
relevance for the self. 

Still another drawback is the child’s deficiency in language. His 
concepts, expressing the relationship between himself and his sur- 
roundings, are but dimly formed, for as yet he lacks the capacity for 
sharply sculpturing thought with words. One illustration: in his “col- 
lective monologues” the two-year-old often confuses quite sadly the 
first, second, and third persons. He may be overheard to say to him- 
self, “You be careful, William get hurt. No! I won’t get hurt.” He is 
first, second and third person all at the same time! It is especially dif- 
ficult for him to learn the correct use of pronouns; he cannot learn 
by imitation, for he has to realize that each pronoun has a different 
reference when used by himself. Although it would not be correct 
to date the dawn of self-consciousness from the correct use of the 
personal pronouns (around two and a half years of age), still the con- 


fusion undoubtedly signifies to some extent the difficulty the child 
is having with his sense of his own individuality. 

Even after self-consciousness is partially established the child read- 
ily surrenders his own identity in play. He may lose it so completely 
that he becomes angry with his parents if they fail to recognize his 
transformations. 3 Even at the age of four or five the self is by no 
means firmly encapsulated. The child continues to confuse himself 
with his surroundings, to take the role of others in play, and to iden- 
tify his private fantasies with objective fact. He does not customarily 
feel himself, as does the adult, to stand out against all that is foreign to 
him, distinguishing in thought and act that which is himself from 
that which is not. 4 

The factors contributing to the growth of self-consciousness have 
often been discussed. 5 During the nineteenth century the account was 
given most often in sensationalistic terms. The infant receives, pre- 

3 This depersonalization of the self by children is called by J. Peres, autoscopie , 
/. de Psychol ., 1926, 23, pp. 558-566. 

4 But even adults in some respects behave as does the child. Most striking are 
those practices among primitive peoples that indicate a merging of the sense of self 
with a consciousness of the surrounding world. The primitive medicine man through 
incantation feels that he “becomes” the rain; the father in couvade identifies him- 
self with his wife at childbirth; injury to one’s personal property causes a feeling of 
illness; one keeps one’s secret name and one’s totemic animal sacred as means of 
self-protection. Parts of one’s bodily self, e.g ., the hair or finger-nails, also are some- 
times treated as equivalent to the whole self. All such practices of identification, 
projection and sympathetic magic indicate that the primitive person, like the child, 
feels himself to be extensible, variable, and capable of merging with events outside 
of his physical body. (Cf. H. Werner. Einfilhrung in die Entwicklungspsychologie , 
1926, Book II, Pt. 4.) 

And even among civilized people there seem to be differences in the degree to 
which the individual closes himself off from his environment. Lewin points out a 
peculiar difference in this respect between the typical German and the typical 
American. Only in the most superficial layers of his personality does the German 
expose himself freely and un-selfconsciously to his environment. He, more often 
than the American, associates his daily activities with his own Ego; he becomes 
emotionally aroused if his theories and ideas are assailed, confides little in his asso- 
ciates, and even keeps his office door closed and his callers waiting in order to en- 
hance his own feeling of pride and worth. The typical American on the other 
hand does not mind if newspapers publish a good deal concerning his private life, 
if people see him at work in his office, if unexpected interruptions deflect him 
from his momentary purposes, or if he is addressed without tide or formality by 
somewhat casual acquaintances. The ego-systems of the American are not so read- 
ily aroused; he is not so instantly and intensely conscious of himself as is the typical 
German. He allows himself to bend and adapt to the pressures of his surroundings 
without feeling as quickly as the German that these pressures are intrusions. The 
latter is more markedly self-conscious (K. Lewin, “Some Social-Psychological Differ- 
ences between the United States and Germany,” Char, <& Pers., 1936, 4, 265-293). 

5 Comprehensive accounts may be found in H. Taine, On Intelligence , trans., 
1889, Bk. 3, chap, i; W. Preyer, Mental Development in the Child , trans., 1893, 
chap, ix; P. Janet, Devolution psychologique de la personnalite , 1929. 



sumably, a stream of organic sensations from the internal organs of his 
body, from his muscles, joints and tendons. This coenesthetic core 
becomes further elaborated with the sensory impulses of touch, taste, 
smell, sight and hearing. The fusion of sensory impressions, particu- 
larly around the kinesthetic sense of postural strain and position, 
originates the sense of self. This theory is wholly congenial to em- 
piricism, to the belief that at birth the infant is a tabula rasa upon 
which the sense of self along with everything else has yet to be en- 
graved. William James found evidence for this sensationalistic theory 
in his observation that for the adult the reportable experience of self- 
hood is usually reduced to a matter of postural strains and stresses, 
centering especially in the head.® Some people, he found, locate the 
self in the facial furrows between the brows, so that midway between 
the eyes the introspective self resides. Following a less sensationalistic 
view, a more abstract location is suggested by Koffka. The self (or 
Ego) is that which lies between right and left , between before and 
behind. The self is the point of reference in all temporal experience 
too, for it lies as well at the exact junction of past and future. 

The conscious recognition of recurring experiences (the sense of 
familiarity) contributes to the development of self-consciousness. 
When an experience is felt as similar to a preceding experience, there 
is always at the same time a vague sense of time-binding, and for the 
individual the person having this conjoint experience of then and 
now can only be himself. 

Then there are certain symbols providing anchorage points for 
selfhood, of which the most important is one’s name. A proper name 
is a mark put upon the individual at birth, first acquiring significance 
for him in the second year of life. The name becomes a more and 
more strategic point of contact between the self and the outer world. 
With the name comes the formality of receiving salutations and ad- 
dress from others, and with this formality comes a sense of self- 
importance and of position within the social hierarchy. The impor- 
tance of this symbol of the self is revealed in the magical practices of 
primitive peoples where the mere pronouncing of a person’s name 
with maledictions is thought sufficient to inflict actual injury upon 
him. Even in civilized societies a man’s good name is to him a sacred 
possession . 7 

« W. James, Principles of Psychology , 1890, II, p. 301. Cf. also, E. L. Horowitz, 
“Spatial Localization of the Self,” /. Soc. Psychol. , 1935, 6, 379-387. 

7 An experimental investigation dealing with the problem of the justifiability of 
homicide, revealed that a man’s honor (/.<?., the regard in which his name is held) is 


To possess a name is only one step in achieving social status. 
Everyone feels more at home ( i.e less self-conscious) in certain 
groups than in others: this is because habitual status in a group pro- 
vides a familiar frame of reference. Outside this frame a person is 
likely to feel uncomfortably aware of himself. The professor speaking 
at a labor-meeting is much more self-conscious than in his classroom; 
the farmer who finds himself in a city drawing room is similarly ill 
at ease. One loses oneself only in surroundings that are familiar and 
in groups where the role one plays is habitual, or else in certain situa- 
tions (a crowd, for example) where each participant is inconspicuous 
and safely anonymous. 

In childhood, clothing, ornamentation, and special grooming con- 
tribute their share to self-consciousness. Investigators have observed 
that little children speak more frankly and with less inhibition when 
unclad. It is as if self-consciousness were a garment as readily shed as 
shoes. Apparently the Nudists in abolishing clothing hope to regain 
some of the child’s freedom from the oppressive burden of self- 
consciousness. Before Hitler officially instituted “morality” into Ger- 
many, Nudist colonies were more numerous there than in any other 
country, perhaps because self-consciousness, as Lewin observed, is an 
oppressive national characteristic. But if it is difficult for the child 
to take on the burden of self-consciousness, it is even more difficult 
for the adult to lay it aside . 8 

Finally, all experiences of pain, frustration, and especially of social 
ridicule engender acute states of self-consciousness that leave perma- 
nent effects. Whenever one is unable to achieve, or to continue in, a 
condition of friendly relation with the environment, he must perforce 
pay attention to his own shortcomings, and thereby become acutely 
aware of the incompatibility between himself and the physical and 
social world outside, and of his isolation. In pleasure, when every- 
thing is going well, this separation is not felt; but pain is always re- 
ferred to the self. 

considered second in importance only to the immediate physical safety of himself 
and of others closely associated with him. Violation of one’s honor is considered a 
far more justifiable cause for homicide than invasion of one’s property and material 
possessions (Cf. G. W. Allport and R. L. Schanck, Char, dr P ers., 1936, 2, 193-203). 

8 Self-consciousness in its popular sense of embarrassment is a hypertrophy of 
the natural awareness of self, intensified by frequent failure and consequent experi- 
ences of shame. It is a common and chronic condition in many personalities. In one 
yudy, one-third of a general population of adults reported it as their main handicap 
in life and chief source of worry (A. A. Roback, Self-consciousness, Self-treated, 
1936, p. 41). 


The advent of self-consciousness in childhood is gradual, and its 
growth continuous, but a certain critical stage is reached around the 
age of two. Its symptom is the period of negativism that is as distress- 
ing to parents as it is interesting to psychologists . 8 Children at this 
period commonly resist persuasion, more often than not disobey, and 
generally protest against interference with their own designs. They 
say “No” much more often than “Yes.” One little boy, not yet three, 
made a daily visit to his grandmother’s house across the street to an- 
nounce (apropos of nothing) , “Grandma, I won’t!” Such impulsive 
contrariness augments the sense of selfhood through the aggressive 
exercise of self-determination. The child regards outsiders as threats 
to his initiative, choice, and freedom of execution. Against all of these, 
prompted by his new and disturbing sense of personal integrity, he 
rebels. A similar counter-suggestibility occurs sometimes in adult per- 
sonalities. Apparently as a matter of principle they immediately dis- 
agree with every proposal or assertion. Spontaneously they say “No”; 
they may later think the matter over to find that they would have 
preferred to say “Yes.” As a mode of preserving one’s integrity such 
negativism is far more appropriate to the two-year-old than to the 

There are l i kewise pathological types of regression and dissocia- 
tion in self-consciousness, cases of psychogenic loss of personal iden- 
tity. In some instances the loss takes the form of a splitting of self- 
consciousness into independent regions, as in fugues and other amne- 
sias; the patient by surrendering his continuous sense of identity lops 
off areas of his life that are distasteful to him, or for which he is un- 
able to assume further responsibility. In completely regressive states 
a patient seems to commit a kind of psychological suicide in order to 
escape entirely from the oppressive burden of some intolerable feature 
of his system of selfhood . 10 

9 The onset of this period is sometimes placed earlier for female and later for 
male children (D. M. Levy and S. H. Tulchm, /. Exper. Psychol., 1923, 6, 304-322; 
1925, 8, 209-224). An experimental study of the stages of childhood negativism is 
described by M. M. Reynolds, “Negativism of Pre-school Children,” Teach. Coll. 
Contrib. to Educ., No. 288, 1928. 

10 Cf. M. Abeles and P. Schilder, “Psychogenic Loss of Personal Identity: Am- 
nesia.” Arch. Neurol. Psychiat., 1935, 34, 587-604. 


1 66 


It would be impossible to estimate how large a portion of anyone’s 
personality is acquired through the instrumentality of suggestion, but 
it must be a very sizable part, because— if for no other reason— a re- 
markably close relationship exists between suggestion and learning 
by means of language. 

Chiefly through speech the child takes over his beliefs and norms 
of conduct from adults. For example, as soon as he is able to under- 
stand and obey he hears from his parent a series of mandatory rules: 
(a) “You must take this medicine,” (b) “You must be quiet while I 
am speaking,” (c) “You must not talk about such things outside the 
family,” (d) “You must attend Sunday School.” Assuming that the 
child has already learned the obligatory significance of propositions 
introduced by the injunction must , he now obeys for the simple rea- 
son that this word has a mysteriously imperative character. The word, 
perhaps, is spoken in an emphatic tone of voice, and may have been 
originally accompanied by physical coercion. Whatever its history 
it acquires with time its unquestioned and irrational sanction. 

The child is unable to see that there are a good many different 
reasons why he “must” behave in prescribed ways. In the case of the 
first of the prescriptions set down above the penalty for failure will 
be a natural punishment, viz., ill health; in the second, the sanction 
is merely parental prerogative based on superior strength and status; 
in the third, the penalty is social ostracism; in the last, divine dis- 
pleasure. Unless, or until, the child by a persistent course of question- 
ing, asks “Why” whenever he is told that he must do thus and so; 
unless, or until, he discovers the answer and considers it a sufficient 
ground for conforming to the course of conduct indicated— unless or 
until such a time arrives, his habits of response will be shaped by sug- 
gestion. If he obeys these four rules unquestionably he will be form- 
ing the rudiments of hygienic, filial, conventional, and religious habits 
and sentiments, all without self-determination, merely by virtue of his 
own suggestibility. By the time he comes of more critical age, decid- 
ing henceforth to guide his own destiny, he is already a creature of 
innumerable conventional forms of behavior and outlooks, acquired 
by suggestion, from which he can never completely escape. 

Suggestion has not as yet been defined. Simply stated, it is the adop- 
tion of a course of conduct or belief by an individual who does not 



engage in those processes of thought and judgment that would be 
pertinent to the acceptance of such a course of conduct or belief. 
More briefly, suggestion is the acceptance of a proposition for belief 
or actio n in the absence of complete self-determination. 

At a rudimentary level, most conditioned responses can be viewed 
as elementary suggestions. An infant, given a nipple detached from 
the nursing bottle, sucks it with considerable show of satisfaction. At 
first the milk in the bottle alone brought contentment, but now by 
frequent association, the outward sign or symbol alone suffices (at 
least until hunger becomes more urgent) . At a more complex level, 
all the verbal pronouncements accepted as gospel truth simply be- 
cause one is accustomed to take words at their face value without 
supplementary evidence or reason, are likewise suggestions. A college 
teacher often finds to his surprise that his lectures, his counsel, and 
even his obiter dicta are adopted uncritically as models of conduct 
and belief by students in whose eyes he has prestige. However benefi- 
cial the influence may be for the student it is none the less a result of 

It is not the case, however, that all the functions of language are 
at the same time instances of suggestion. Words may in fact be sharp 
instruments for reasoning and as such aid in die formation of self- 
determined concepts. Yet, even when used with the maximum of 
critical discrimination a word inevitably carries with it the weight of 
pre-existing socialized forms of thought. The result is that the thinker, 
as he comes to depend on verbalized concepts, tends more and more 
to guide his conduct and to build his personality into conventional 
molds. Perhaps one chief reason why personalities within the same 
culture resemble one another as much as they do is the common pos- 
session of linguistic symbols that give common meanings, common 
evaluations, and common guidance to the thought and to the conduct 
of the various members of any one cultural group. 

We cannot say at what age suggestion plays its chief part in the 
development of personality. The younger child, in spite of his neg- 
ativism, is really unable to resist the weight of authority of those 
who teach him, and takes on his earliest social behavior almost en- 
tirely by virtue of his unwitting suggestibility. Somewhat older chil- 
dren, around eight or nine years of age, owing to the rapidly advanc- 
ing period of linguistic development, are especially suggestible; for at 
this age a vocabulary is built of moral, religious, political, and esthetic 
concepts— all extremely significant for the development of person- 


1 68 

ality. Again, the college age, in some respects, is even more sug- 
gestible, for at that time submissiveness to prestige and to the printed 
word has reached, through long training, such a peak that students 
sometimes seem completely to lack minds of their own. Nor are 
adults in their mature personalities free from the effects of suggestion: 
witness the role of propaganda in shaping their political, moral, rec- 
reational conduct, and its effects upon their habits of purchasing, diet, 
travel, dressing, investing, war-making, and home-making. 

Since suggestion is so significant an aspect of personal develop- 
ment it becomes important to determine whether some people are by 
nature more suggestible than others, that is, whether suggestibility is 
a trait existing in some fixed amount in each personality. This is a 
much-investigated question, and the evidence on the whole is nega- 
tive. To be sure, a few individuals seem chronically to accept any and 
every suggestion offered them. Lacking the power to resist proposals 
that are discordant with their own self-determined plans of action, 
they give themselves up to the situation, trusting that in so doing, 
things will go well, as perhaps they have in the past when the same 
surrender was made. But for most people, suggestions, if habitually 
accepted, would be ruinous to the integrity of traits and ideals already 
established. Usually each of us is suggestible in certain ways, in direc- 
tions where we have already a strong desire to believe or act in the 
suggested way, or in directions where we lack knowledge and convic- 
tion. By its very nature suggestion involves only part of the person- 
ality; it is a species of dissociation, operating only when resistances 
are weak. It is a capacity that all people have, but only in a few is it 
a well-integrated disposition to adjust in a positive way by passing 
all decisions over to some outside control. In other words it is rarely 
a trait. 

A similar answer must be given to the question, “Is negativism a 
trait?” In a few people, yes. On p. 165 we mentioned certain adults 
who seem, like the typical infant of two or three years, to be always 
on guard, resisting every proposal on principle for fear they may 
think thoughts or perform acts contrary to their own natures. Such 
contredismt individuals there are, indubitably possessing a trait of 
negativism. But ordinarily, each of us is negativistic only in specific 
ways; we feel no positive urge to contradict every proposal that 
comes to us: only those propositions that offend firmly established 
sentiments and beliefs, or that violate mature traits, are rejected. 




All the philosophies of Egoism, and many others as well, stress 
the demand for self-aggrandizement in human nature . 11 Not hing , it is 
said, is ultimately sacred excepting the beloved Ego. Motives ordi- 
narily regarded as self-sacrificing and other-regarding are at bottom 
merely selfish. Scratch ever so lightly the coating of hypocrisy, of 
social varnish, and the cave-man stands revealed. The one important 
instinct is the desire for power, for “masculinity,” and though over- 
laid with sweet-sounding protests of sympathy and altruism, this root 
desire is biologically prepotent and ultimate. Every man is inescapably 
a Machtmensch; his most coveted experience is the enhancement of 
his self-esteem, and his most ineradicable trait is vanity. 

Such a doctrine of human nature contains so much obvious truth 
that it is often accepted uncritically as a fully adequate interpretation 
of personality. The next chapter will show that socialization is not 
simply a varnish laid over personality, but involves, at least much of 
the time, a genuine transmutation of interests from the egoistic to the 
altruistic. The biological creature that we find in early childhood pos- 
sesses no instincts, habits, nor sentiments that are in the remotest de- 
gree socialized or civilized. Egoism is the incontrovertible philosophy 
of early childhood. But in the process of growth and extension of 
interests, newly adopted codes and manners represent genuine, not 
superficial, alterations in personality. Nor is the transformation merely 
one from unenlightened self-interest to enlightened self-interest. 
Demonstration of this point must wait. In the present section due 
weight and consideration shall be given to the importance of self- 
esteem in directing the development of personality. 

When an adult undertakes to perform a task he generally places 
his goal at a level not so far above his abilities that he will suffer embar- 
rassment and humiliation if he fails, nor so far below his abilities that 
he will feel ineffectual and cheap upon accomplishing the task. He 
undertakes that amount and that kind of labor which will keep his 
self-esteem at a maximum . 12 It is true that some people prefer to make 
certain of success, and accordingly undertake no more than they can 
surely achieve. Others, more characteristically daring, bite off more 

“For example, F. Lc Dantec, L’Egohme , 1918; likewise the philosophies of 
F. Nietzsche and M. Stirner. 

12 F. Hoppe, Psychol. Forsch., 1930, 14, 1-63. 


'I 1 . 



than they can chew, maintaining their self-esteem either through this 
act of courage or through the closeness with which their accom- 
plishment corresponds to their ambition. They like to be surprised 
that they have approximated the goal as closely as they have. 13 But in 
every case one’s level of aspiration betrays in some way the “upward 
tendency of the Ego.” 

It is likewise known that younger children ordinarily prefer to re- 
peat tasks in which they have already succeeded, unlike older chil- 
dren and adults who prefer to work on tasks as yet uncompleted. 
Feeling that they will be humiliated if they fail to accomplish their 
goal, older persons persevere, while younger children avoid humilia- 
tion by demonstrating over and over again their success on a low 
level of accomplishment and leave difficult tasks uncompleted with no 
signs of embarrassment. The older person battles against outer reality 
to retain his self-esteem; the young child in his world of pleasure pre- 
fers to hold to his earlier and assured successes. 14 

These experimental studies all seem to bear out the traditional 
dicta of philosophers: “The deepest principle of human nature is the 
desire to be appreciated”; “Self-defense is nature’s eldest law”; “By 
whatever name we call the ruling tyrant, Self is all in all.” The center- 
ing of each life upon its own sense of integrity and self-importance 
is everywhere recognized. In psychology, Freud’s concept of Nar- 
cissism has found a prominent place. Koffka postulates as a paramount 
principle of dynamic psychology “a force which propels the Ego 
upward.” 15 McDougall has found at the heart of every personality 
the central sentiment of self-regard, playing “the most powerful all- 
pervasive role in the higher life of man.” 16 

Now, with what are we dealing here? Is self-esteem a force, an 
instinct, a sentiment, or what? Is it not perhaps a self-evident prin- 
ciple, a psychological redundancy, equivalent to that vague, though 
unquestionable primordial “will to live,” postulated on page 109? 

13 J. D. Frank, Anier. J. Psychol., 1935, 47, 119-128. 

14 S. Rosenzweig, ]. Genet. Psychol., 1933, 42, 423-441. 

If> Principles of Gestalt Psychology, 1933, pp. 670 f. 

16 Energies of Men, 1933, p. 234. See also, An Introduction to Social Psychology, 
1908, chap, vii, and Outline of Psychology, 1923, pp. 426-434. This writer believes 
that “self-regard” is the best name for the sentiment, whereas “self-respect, self- 
esteem, self-love, pride, ambition are the names of distinctive types of self-regard. 
Selfishness, egoism, egotism, vanity, conceit, humility, megalomania, swelled-head, 
bumptiousness, pushfulness, masterfulness, aggressiveness, these are some of the 
qualities of personality determined in the main by the composition and mode of 
working of the sentiment.” Energies of Men, p. 233. 



Considered in this way the psychologist cannot make use of it as an 
explanatory principle. To do so would be to beg the question. It must 
be broken open and studied in its manifold operations. One cannot 
treat self-esteem or self-regard as an entity; for basically co-extensive 
with life itself, it enters into all sentiments and traits, which are 
after all merely channels of the primordial (non-psychological) life- 

We must adopt a more discriminating view of the problem. Let 
egoism with its conscious accompaniment of self-esteem be admitted 
as an initial principle of life, especially manifest in the early behavior 
of children. Let the numerous aspects of growth discussed in these 
chapters represent so many ways of focalizing, channelizing, and 
radically redirecting this original “metaphysical” flow. Then, in order 
not to overstate the case for transformation, let it be admitted that in 
spite of all the alterations that egoism may undergo in the course of 
development, there frequently remains at the core of self-conscious- 
ness a strong element of self-seeking and vanity, which likewise may 
be traced in many, perhaps most, of an individual’s sentiments and 
traits. The task of the psychology of personality is to characterize all 
the innumerable and variable contexts in which the element of self- 
esteem occurs, including those where it is no longer a crude factor, 
but is drastically altered and transformed, as well as those where its 
operation is as yet unsocialized and primitive. 

Whatever the ultimate character of this principle, its cruder forms 
of expression result in extraordinary strategies of conduct. It alone 
is responsible for a great super-structure of masquerade built up in 
every life. All in the interests of self-esteem one may cover one’s 
true emotions, put on a front, and at considerable cost avoid exposing 
one’s weaknesses. The persona that develops protects one from unwel- 
come narcissistic wounds. 

What is even more spectacular, likewise in the interest of self- 
esteem, is the capacity men have for deceiving themselves. At first 
sight the ability to fool oneself would seem to be a fatal invention 
of nature, for living would seem to require an accurate evaluation of 
one’s own motives and capacities. Why should an intelligent person 
invent an eye-wash for himself? Because an eye-wash, surface treat- 
ment though it is, brings immediate relief, preventing conflicts from 
developing through the sense of being in the wrong, and engendering 
a certain bravado necessary for life, and for maintaining one’s rights 
in the face of immediate opposition. Self-deception also enables one 


for the time being to put off the admission of unpleasant truths until 
one is ready to receive them. 

The techniques of self-deception are numerous . 17 Psychological 
usage groups them all under the single rubric rationalization, a term 
signifying, of course, precisely the opposite of reason. Reason may be 
defined as one’s capacity to shape one’s belief and conduct to accord 
with one’s knowledge of the world, and if one’s knowledge is in- 
sufficient, the capacity to set out to acquire more knowledge perti- 
nent to the issue in hand. Reason fits one’s impulses and beliefs to the 
world of reality; rationalization fits one’s conception of reality to 
one’s impulses and beliefs. Or, as the aphorism has it, reasoning dis- 
covers real reasons, and rationalization, good reasons, for what we do. 

Rationalizations range from the trivial to the grandiose. Not in- 
frequently we do something on impulse and then call it by the best 
name possible. As Emerson has said: “That which we call sin in others 
is experiment for us.” At the other extreme one finds elaborate sys- 
tems of “metaphysical absolutes” built up to justify tenacious convic- 
tions; since these convictions would be held even in the face of plenti- 
ful contrary evidence, rationalization enters to make them seem as 
reasonable as possible. According to Lotze, a man’s philosophical 
creed is more often than not merely an attempt to justify a funda- 
mental view of things adopted once for all early in life. The same 
thought underlies Pareto’s doctrine of derivations. But one must be 
careful. The reduction of all philosophical activity to mere rationali- 
zation is dangerous; for underneath there must be an ontology of the 
Being who rationalizes, and a logic capable of distinguishing rationali- 
zation from true reasoning. It gets us nowhere to say that all philoso- 
phy is a rationalization of what the private life of the philosopher 
secretly holds. For to reach such a cynical conclusion it is necessary 
to trust one’s own reason as well as the canons of logic. 

Quite a different example of rationalization is that special form 
known as projection. It may be defined as a type of self-deception by 
which a person ascribes his own secret thoughts, wishes, and short- 
comings to another person. If one can castigate others, one is thereby 
saved from the painful duty of castigating oneself. “It is not hard to 
observe,” writes Goethe, “that in this world man feels most free from 
his sins and most blameless when he can comfortably expatiate on the 
same shortcomings in others.” There is likewise a complementary 

A valuable classification and discussion are offered by R. C. Cabot, The 
Meaning of Right and Wrong, 1933, pp. 283-347. 


form of projection whereby a person does not attribute his own frame 
of mind to others but rather one that explains and justifies his own 
frame of mind to himself. Thus the over-timid child thinks that 
others have aggressive intentions toward him, and the paranoiac be- 
lieves that others are plotting his destruction. Through such comple- 
mentary projection one’s personal quirks of temperament and traits 
receive “rational” explanation, and do not appear as unfounded and 
foolish as they are. 

And so there are many, many methods, some direct and some in- 
direct, for keeping self-respect and self-esteem at the highest possible 
level. In ordinary undertakings the level of aspiration automatically 
adjusts itself to serve this purpose. When the direct achievement of a 
high level of self-respect is not possible, roundabout tricks are re- 
sorted to: the persona is donned, defenses spring up, rationalizations 
become rife, and projections are unconsciously devised. But the most 
interesting of all the sly handmaidens of self-esteem has not yet been 
described— the principle of compensation. 


The successive rhythms of maladjustment and adjustment consti- 
tute the pulse of development. Some intrusive factor, perhaps hunger, 
a change in temperature, or some social demand, upsets an insecurely 
established equilibrium. There are resulting tension and unrest, fol- 
lowed, through the instrumentality of higher mental functions, by an 
attempt to restore the equilibrium. If a straightforward adjustment is 
possible the problem is met and solved, at least temporarily. And if a 
direct solution is unsuccessful and variations in the method of attack 
do not succeed, the failure sometimes is minimized, repressed, or 
rationalized out of the way. 

Often, however, when failures are recurrent and serious they can- 
not be so easily disposed of. A tension not relaxed by fulfillment is 
present in a latent state and always ready to cause trouble whenever 
the desire for the unattainable goal returns. As a result a deep-seated 
sense of deficiency may develop and be steadily aggravated. The 
sense of deficiency may be due to different causes: physical incapac- 
ity, ill health, low vitality, sexual impotence, unpleasant appearance; 
to social inadequacy— poverty, lack of education, gaucheness, or lack 
of wit and self-possession; to faulty intelligence— poor memory, 
meager vocabulary, deficient accomplishment; or to moral conflict— 

T 74 


a sense of unworthiness, of guilt or of sin. As failures multiply, the 
source of difficulty becomes the focus of attention and concern. The 
sufferer feels habitual uncertainty or fear in the face of those situa- 
tions that threaten to reveal to himself and others his own weakness 
and ineffectuality. This condition is the famous inferiority complex. 
If a definition is required, one may say that the inferiority complex 
is the strong and persistent tension arising from a somewhat morbid 

Type of inferiority feelings 





Percentage reporting persistent inferiority 






















l 7 



none at all 





A Table of “Inferiority Complexes” 
Among College Students 

emotional attitude toward one’s failure to effect a satisfactory direct 
adjustment to his environment, owing to some felt-deficiency in his 
personal equipment. 

Few people need to have explained to them the discomfort caused 
by feelings of inferiority. In one study reports from college students 
were secured concerning the four types of inferiority feelings de- 
scribed above. 

Less than ten per cent of the students report that they do not 
know what it is to suffer from gnawing feelings of inferiority. The 
table shows that as students grow older they show a tendency to 
suffer less from a sense of inferiority. On the whole the women seem 
to be somewhat worse off than the men; not only do they report a 
larger number of such feelings, but also they show a less marked 
reduction in conflicts with growing maturity. In one of the categories, 
the intellectual, the inferiority feelings of the women have markedly 


increased, and are far more numerous than among men . 18 In both 
sexes the sense of social inadequacy seems to lead, but this category 
is not definitive, since almost any type of inferiority is reflected in 
social inadequacy. Half the students are afflicted with a sense of 
physical inferiority, and it was, in fact, the recognition of the role 
of such organic deficiency that led to the first psychological recogni- 
tion of the complex by Adler . 19 Moral inferiorities are least frequent, 
probably not because of the superior virtue of the students, but be- 
cause of the diminished emphasis in their generation upon moral 
lapses and sin. 

Hidden within this table is one conclusion of importance. Feelings 
of inferiority cannot be taken as cm index of actual inferiority. One 
notices, for example, that over one-half of the students have at one 
time or another suffered a sense of intellectual inferiority, an absurd 
situation from the factual or objective point of view. Over half the 
group cannot be below the average in intelligence; statistically they 
cannot be inferior. But more to the point, college students, it is 
known, are actually superior in intelligence to the average population. 
Yet they suffer because of their shortcomings in mental ability. It 
is also known that college students are by and large superior in 
physique and health, and that they have better than average social 
and economic advantages; and still they suffer. Inferiority feelings 
obviously are not based on factual inferiority but are subjective 
phenomena, engendered entirely by the ratio that obtains between 
success and aspiration. The second best chess player in the world 
might suffer miserably from feelings of inferiority, so too might the 
topmost scholar in a university whose aspirations were in excess of 
his great ability. As to physical inadequacy, who is not a candidate 
for an inferiority complex unless he is that rara avis, a biological mil- 
lionaire? But if he were a biological millionaire he might still have 
inferiority feelings of a moral order because, unlike his less favored 
associates, he had never experienced physical suffering! 

To return to the question of sex differences. The higher ratio of 

18 This curious result may have only local significance. The women in the study 
were students in a girls’ college in the shadow of a large university attended only 
by men. They were dependent for all their instruction on teachers from the neigh- 
boring university, and the fact that ail of the teachers were men seemed to under- 
mine the women’s confidence in the intellectual abilities of their own sex, and of 

19 A. Adler, Organminder'wertigkeh mid ihre psychische Kompensationen , 1912: 
also trans., 1917, Nerv. & Menu Dis . Monog, Series , No. 24. 



inferiority feelings among women no doubt reflects the disadvantage 
they feel in a “man’s world.” Over and above whatever handicaps 
they have as individuals they have extra restrictions placed upon 
them, especially in economic and moral spheres of activity. The fol- 
lowing table sheds some light upon this problem. It is based upon 
anonymous replies from four groups, about 300 cases in all. 20 

Percentage Replies 



W or king 






Have boys on the whole an 






easier time in life than girls? 





! 9 

Did you ever wish to be of the 






opposite sex? 





8 l 

Would you now like to be of 


: 30 




the opposite sex? 


1 70 




Table of Attitudes Toward 
Opposite Sex 

The majority of all groups regard the position of growing boys 
to be more enviable than that of girls. Nearly three times as many 
girls as boys at some time have wished to be of the opposite sex. Al- 
though most of the young women are now content with their role, 
the desire to be a male is by no means uncommon among them, espe- 
cially in the population of working girls. The corresponding desire 
among boys is apparently non-existent. How much this last result 
reflects the victory of the Super-ego, preventing an honest report, 
and how much the general conflict on the part of women reflects a 
“masculine protest” in the sexual sense, cannot be decided. But what- 
ever part these or other unconscious factors play in the results, their 
significance for our purposes is unaffected: girls have a harder time 
to make satisfactory positions for themselves in their environments, 

20 The results were obtained with the kind assistance of Miss Theresa Larkin 
and Mr. W* H. Clark. 


and for this reason suffer more commonly feelings of inferiority. 21 

What does one do about an inferiority complex besides suffer 
from it? The tension it engenders, by its very nature, defies simple and 
direct methods of alleviation. One cannot permanently repress it, nor 
can one always escape it by “going out of the field.” Some more 
sustained form of combat is required, and to this form Adler has 
given the name, compensation. 22 

Many forms of compensation have been described in psychologi- 
cal literature. 22 Only a brief characterization of each need be given 

Direct action, or compensation in kind, occurs when through 
persistent attack upon the source of an actual inferiority it is finally 
for all time removed. When, as occasionally happens, the source of 
the deficiency is not only removed but actually converted into a 
source of strength, one speaks of over-compensation. Demosthenes, 
legend has it, worked so persistently to overcome his stammering 
that he became not merely a normal speaker but a great orator. Theo- 
dore Roosevelt whose early frailty caused him to suffer feelings of 
inferiority, built up his physique through systematic training, but in 
so doing he owr-compensated, and emerged a lion-hunter and a 
roughrider (not merely a partridge hunter and a smooth rider). 
Everyone knows of immigrants who upon arrival in the United States 
felt themselves to be socially despised and inferior, but who by at- 
tacking directly their difficulties rose to remarkable heights of success. 
The American classic of the self-made man is nearly always a story 

21 In the home, the only niche allotted them in previous times, women did not 
have such exaggerated feelings of inferiority. It was their world, freely granted 
them by men. Few of the sex left the niche to compete in the “world of men.” 
Nowadays it is the world of men in which many women are living and competing, 
a world of standards intrinsically alien to them as women. Slowly those standards 
are being modified to include them. As this occurs, as women are admitted on equal 
terms, the ratio of inferiority feelings may be equalized. No doubt the change will 
also affect the percentage of girls preferring to be boys. 

22 The psychological significance of this term is much narrower than its com- 
mon usage. It is not the cosmic principle that Emerson would have it, nor is it 
synonymous with every attainment of equilibrium and adjustment. When, for ex- 
ample, a short person walks faster to keep up with a tall person, he is, in common 
speech, compensating for his inferior height, but psychologically he is in all prob- 
ability merely adjusting his gait. But if at the same time he is telling tall tales of 
his own prowess to his companion he is probably compensating for a chronic in- 
feriority complex now re-aroused by the contrast in heights. 

23 The gist of Adler’s teaching' filtered rapidly through the institutes of psy- 
chology into popular lore, psychologists assisting in the popularization. Cf. R. Dodge 
and E. Kahn, The Craving for Superiority, 1931, and W. F. Vaughan, The Lure of 
Superiority, 1928. 


of compensation in kind. It goes without saying that the mental hy- 
gienist regards such compensation as the most desirable form both 
from the social and the personal points of view. 

Substitution. Often the source of difficulty cannot be removed 
by compensation in kind. One must seek satisfaction in an entirely 
different direction. The hunchback cannot correct his deformity, but 
he may, through application or cunning, become the power behind 
a throne. The plain girl may develop a compensatory charm and wit, 
or the non-athletic youth through diligence may excel at his studies. 
As a young man the philosopher Immanuel Kant could not reconcile 
himself to his sunken chest, restricting the free action of the heart 
and lungs. He describes how this condition predestined him to hypo- 
chrondria from which he suffered greatly. Little by little as it became 
clear to him that nothing could be done to change his physique, he 
centered his attention more whole-heartedly upon his stronger abili- 
ties. In his own words, he taught himself to be “calm and clear in the 
head, although oppressed in the chest.” 24 

Defense Mechanisms are compensations designed to deceive others. 
Many of them are of a rather trivial order, mere habits or mannerisms 
calculated to throw others off the scent. One can often recognize 
defensiveness in the exaggerated handshake of a self-conscious adoles- 
cent striving to mask his own embarrassment, or in the bluster of a 
bully who throws a smoke screen around his own weakness, or in 
goatees, thick-soled shoes, disarming smiles, or ingratiating manners 
affected in order to hide physical defects of stature or feature. More 
highly organized defense mechanisms are those typical of the brag- 
gart, the pathological liar, and the person who “protests too much” 
in the effort to hide feelings of inadequacy or guilt. 

Self -justification and Rationalization are forms of compensation 
unconsciously designed not only to fool others but likewise to fool 
oneself. When failure occurs, however minor it may be, the first 
reaction often takes the form of self-justification. The beloved Ego 
must be allowed to triumph; if it cannot do so in a realistic encounter, 
it may do so in retrospect. How often conversations overheard in 

24 One might, somewhat less securely, go further in Kant’s case and point out 
that not merely the fact of becoming a philosopher, but the nature of his philo- 
sophical teaching bears traces of the compensatory urge. He had failed in the emo- 
tional field; that is to say, most of his emotions had been unpleasant ones of suffer- 
ing and defeat. He substituted, therefore, an emphasis upon reason., pure and 
practical, at which he himself so greatly excelled, comforting himself with the con- 
viction that emotions, after all, were “diseases of the intellect.” 


the street car reduce to the simple formula: “He said to me . . . but 
/ said to him!” Watch the wagging assent of heads. There is never 
any question as to who was right in the argument. In cultivated per- 
sonalities of good insight or social training this impulsive self -justifi- 
cation is less often encountered (Chapter VIII). 

When one slips constantly below a standard of conduct chosen 
for oneself (in morality or in the display of strength, wit, or grace) 
one usually finds extenuating circumstances to minimize the force of 
the failure. Sometimes the extenuation takes the fonn of a permanent 
protective rationalization. A pale non-athletic young man defended 
(and deceived) himself with the comforting sentiment, “I am sick of 
hearing about red-blooded athletes. I have decided what red blood 
means; it is the blood that never flows through the brain.” One man 
who admitted that he was by nature slow, added for his own comfort, 
“but sure.” Another with an especially cadaverous face thought it not 
such an affliction since it made him distinguished in appearance “like 
Rameses or Savonarola.” The youth who decided that what he 
couldn’t have wasn’t worth having was indulging in sour-grapes 
rationalization; the man who looked like Rameses and the other who 
though slow was sure, were making virtues of their respective neces- 
sities in a sweet-lemon rationalization. 

To accuse oneself, or to be accused, of dereliction and incompe- 
tence causes tension and annoyance. Better, if possible, to find an 
“alibi,” an immediate excuse, and so close the subject. Often such a 
stop-gap proves adequate; suspicion of incompetence is lulled; tension 
is removed. It is to be sure only a pars pro toto closure, and may not 
—like direct action— permanently defeat the inferiority feeling. Sooner 
or later some more thoroughgoing type of compensation may have 
to be instituted. 

Rationalization has its institutional aspect. A great many people 
with social and intellectual inferiority feelings find solace in the 
sour-grapes and sweet-lemon attitudes of journalism. They read com- 
forting news of the stupidity of the “brain trust,” the evil ways of the 
elite, and the uncouthness of college students. In pulp magazines and 
on the screen they find the homely virtues of poverty and semi- 
illiteracy extolled. To provide such ready-made rationalizations is one 
duty of the fourth-estate, committed, as it is, to giving the public 
what it wants. This phenomenon of institutional rationalization pro- 
vides one of the principal problems of social psychology. 

Autistic Thinking ( fantasy ) is compensation that occurs when 



an individual disregards completely the demands of his physical and 
social environment, withdrawing into himself to day-dream of suc- 
cess. A certain slim youth suffering persecution from his tougher 
schoolmates, fled to his room every afternoon after school to play 
his two favorite games. In one he was a schoolmaster, the head of a 
large class of boys (his own tormenters) . In this role he wrote out 
severe assignments and administered floggings to his heart’s content. 
In the other game he was an English country gentleman. Living in 
an imaginary hunting lodge he entertained titled guests and wrote 
fabulously large checks for his favorites. 

Personalities having a marked fantasy life are introverted. When 
the process is carried further so that the outer life, conducted in a rou- 
tine way, has little connection of any kind with the inner life of 
memory, imagination, and wish, one speaks of a schizoid personality . 28 
Whereas to his associates the ordinary introvert may seem merely 
imaginative, an individual with such a marked cleavage seems defi- 
nitely queer. Still more extreme cases are schizophrenic — pathological 
to such a degree that no one can tell what curious forms the imagi- 
nation is taking, and to what delirious unreality the patient has sur- 
rendered himself. 

Neurotic Compensation. Anyone who is ill cannot be expected to 
work energetically, to expose himself to disagreeable social duties, to 
succeed in competitive sports. In general the ill person has an accept- 
able excuse for, and protection from, all failure. Because this is so, 
there are conditions of purely psychic invalidism (usually neu- 
rasthenic or hysteric) that represent the last trump for otherwise in- 
solvable inferiority complexes. A pseudo-malady is the last line of 
defense, and represents so serious an imbalance, so complete a defeat 
for the normal mode of life, that it belongs more properly in the field 
of abnormal psychology. 

We see therefore that the forms of compensation range from the 
most deliberate and persevering attempts to remove the source of 
difficulty by compensation in kind, through a whole course of tricks 
involving self-justification and defensive maskings, both fraudulent 
and fanciful, to down-right pathological surrender. Goals that cannot 
be achieved through the direct channels of adjustment, are pursued 
through one compensatory bypath or another. 

Like suggestibility, compensation is a basic potentiality of the 

25 An unusually good case of this type is that of M. S. in G. H. Green, Psycho- 
analysis in the Class Room s 1923, pp. 33-40. 


1 8 1 

human mind. Everyone is capable of using it. Like suggestibility it is 
not often a distinctive trait. A few personalities, to be sure, are so 
colored by their painfully evident, systematic striving to overcome 
their handicaps at all costs and by whatever method possible, that 
one may speak of a trait of compensation . 20 In these cases there is 
an oddly tenacious, positive quality in the program of compensation. 
But as with suggestibility, only rarely may compensation in a per- 
sonality be identified as a trait, but it is very often a device through 
which traits develop. 


The tremendous vogue of psychoanalytic interpretations of per- 
sonality may perhaps be explained somewhat as follows. Whenever 
there is a striking lack of proportion between an act and the apparent 
reasons for it, there is a presumption that some hidden and uncon- 
scious impulses guided the act. Any psychological doctrine that offers 
a consistent account of these impulses and their operation promises 
thereby to amend all the shortcomings of traditional intellectualistic 
psychology. Before Freud there was in general a fatal neglect of im- 
pulsive emotion and its subterranean workings. What Freud did was 
to insist that the neglected facts of emotion are the most important 
facts of all for psychology. The instant promise of such a new dy- 
namic doctrine attracted doctors, laymen, and psychologists them- 
selves, dissatisfied with the older intellectualistic outlook. Those who 

26 Such a oase is described in the personality of G. L., J. Abnorm. & Soc. 
Psychol., 1921, 16, 6-40* 

27 Devotees of psychoanalysis will no doubt be distressed to find here so tardy 
and so incomplete a review of the contributions of Freud and of his many disciples, 
both orthodox and dissident. There are three reasons why the account is so critical 
and so brief, (i) Psychoanalytic concepts are drawn exclusively from neurotic and 
pathological material, i.e., from cases where imbalance prevails over balance, and for 
this reason their applicability to normal personality is in many respects questionable. 
(2) The portions of psychoanalytic doctrine most valid for normal personality are 
incorporated elsewhere in these chapters, in contexts unfamiliar to psychoanalytic 
theory, but more suitable. (3) The story of psychoanalysis is too familiar to require 
another detailed exposition. One cannot improve upon the accounts of Freud, 
A General Introduction to Psychoanalysts, trans. 1920, New Introductory Lectures 
on Psychoanalysis, trans. 1932; of G. Murphy and F. Jensen, Approaches to Per- 
sonality, 1932; or of I. Hendrick, Facts and Theories of Psychoanalysis, 1934. 

Such a volume as F. Alexander’s The Psychoanalysis of the Total Personality, 
trans. 2930, wrongly implies that psychoanalysis is equipped to deal with the whole 
of personality. The truth is that it deals only with a fraction of the phenomena 
encountered in a comprehensive study of the subject. But in spite of its narrowness 
the bulk of all literature on the psychology of personality is written from this one 
point of view. It is time that the story be told in more eclectic terms! 



wanted a dynamic psychology as guide had no other choice; it was 
a question of psychoanalysis or nothing. There have been, to be sure, 
min or and even major departures from the initial teaching of Freud, 
but broadly speaking, ever since its epoch-making formulation, psy- 
choanalysis has served as the one and only pivot for all thorough- 
going dynamic psychology . 28 

In spite or its subtle secondary elaborations, the Freudian picture 
of personality is essentially simple. Freud does precisely what Plato 
and a host of faculty psychologists have done— he divides personality 
into three arbitrarily conceived parts. The special names for these 
divisions are the Id, the Super-ego, and the Ego, which roughly may 
be translated as Emotional Impulse, Conscience, and cognitive Self- 
consciousness, respectively. The last of these is without energy of its 
own; from the dynamic point of view it is weak. As the merely pas- 
sive principle of conscious selfhood it is continually buffeted by 
three “tyrants,” the objective world, the Super-ego and the Id. 

Not infrequently the Ego, perceiving its own weakness, breaks 
out in distress, suffering vague or specific feelings of fear, i.e., an 
“anxiety neurosis.” (Anxiety, though obviously not a universal trait 
among normal people is a common condition among neurotics, and 
may be said to be the raison d’etre for the whole theory of psycho- 
analysis) . The mission of psychoanalysis is to strengthen the Ego by 
making it more cognizant of the forces of the Super-ego (the social- 
ized conscience of which it is already partially aware) , and of the Id 
(the unconscious store-house of instinctive impulses). If psycho- 
analysis can widen the Ego’s field of vision, so that it is able to take 
over into the unified region of consciousness many hitherto hidden 
portions of the Id, reconciling them in the process with the demands 
of the Super-ego and the outer world, the patient will then face life 

28 Another explanation of the spread of psychoanalytic theory is its dramatic 
and imaginative character. It presents the problem of good and evil dramatized in 
new terms. In its simplicity it can be understood by the layman; he can participate 
in the objectified drama of his own conflicts which has great therapeutic advantage. 
It is a fresh approach, free from the religious terms which as likely as not he has 
repudiated (probably because his own moral growing pains have been associated 
and entangled with religious sanctions and restrictions) . He can accept these new 
terms without losing caste with himself or his contemporaries. Witness how often 
the conversation of a “convert” to psychoanalysis is saturated with the terms of 
the doctrine much as the letters of our grandmothers were with a Biblical vocabu- 
lary. One might speculate as to whether a child brought up in the rigorous pattern 
and terminology of psychoanalysis might not, faced with severe conflicts in adult- 
hood, find fresh and helpful vigor in the dramatic terminology of good and evil 
of the older religions, much as the reverse situation holds today. 



more serenely, and his neurotic difficulties will presumably vanish. 
Psychoanalysis, says Freud, aims primarily at the reclamation of the 
Id by the Ego . 29 Though the goal is therapeutic, the originality of the 
doctrine extends beyond the practical domain into the region of 
theoretical psychology as well. 

Its theoretical significance lies largely in the specific mechanisms 
that are postulated to account for the various relationships obtaining 
between the Id, the Super-ego, and the Ego. These mechanisms psy- 
choanalysis describes in great detail. And yet the account often seems 
to the impartial reader altogether exceptional or else badly exagge- 
rated. Derived as they are from the inductive study of unbalanced 
(anxious) personalities, they are not able, taken collectively, to pro- 
vide a well-proportioned account of the normal course of develop- 
ment. Yet some of the mechanisms described, provided they are kept 
in perspective and not regarded as the alpha and omega of genetic 
psychology, have value for the study of normal lives. Elsewhere in 
this volume, in contexts where they are useful to the normal person- 
ality, some of these are employed (for example, rationalization, pro- 
jection, fantasy, infantilism, regression, dissociation, trauma, the com- 
plex, and the ego-ideal, as well as the unique methods of investigation 
employed by psychoanalysis). A few other concepts must be added 

The pleasure principle emphasizes the fact that immediate grati- 
fication, regardless of future consequences, is normally the demand 
of impulses that are not controlled by an organized and mature Ego. 
The pleasure principle is most apparent in the unsocialized and hedon- 
istic behavior of the young child. The reality principle, a complemen- 
tary concept, is the control that the Ego acquires over the pleasure 
principle through restraining habits and sentiments that are inevitable 
products of adaptation to the moral, social, and physical environ- 
ments. The reality principle is simply a short-hand designation of 
the enormously complex process of the maturing of personality, of 
its fitting into a socialized and civilized setting. The two following 
chapters offer a more detailed analysis of this process. 

The imconscious is an inclusive abstraction, referring to all the 
operations fashioning personality without the individual’s direct 
knowledge. Psychoanalysis tends to populate the unconscious with a 
more or less standard equipment of mechanisms and content. Its 

28 S. Freud, New Introductory Lectures on Psychoanalysis, 1933, p. 112. 

184 the self and its constraints 

primary function is to serve as a store-house of the impulses rejected 
by the Ego. These impulses are rejected by repression , the process 
whereby the individual denies conscious and overt outlet to unwel- 
come or unsocialized wishes. “Everything contradictory to the ruling 
tendencies of the conscious personality, to its wishes, longing and 
ideals, and everything which would disturb the good opinion one 
likes to have of oneself is apt to be repressed.” 30 Repression is not 
only a result of conscious conflict (cf. p. 1 19), but in turn breeds un- 
conscious conflict, with dire symptoms that creep unwittingly into 
speech and behavior. One of the symptoms of repressed emotion is 
symbolization, the representation of unconscious thoughts in accepta- 
ble forms in dreams, art, metaphor, wit, or folklore. Similar is dis- 
placement, by which the unconscious wish is admitted to conscious- 
ness only after some acceptable distortion occurs, so that the true 
and original object of the impulse is supplanted by a surrogate object 
(the hated father, perhaps, becomes an ogre or an executioner). 

Underlying all such mechanisms is the primary fact of conflict. 
Impulses may be antagonistic to each other (e.g., the “Life” and 
“Death” wishes), or the wish may be in conflict with repressive 
standards (the Super-ego) or with the direct requirements of the en- 
vironment. Ambivalmce is a frequent consequence of conflict, the 
same object receiving both love and hate, aggression and fear, accord- 
ing to its capacity for arousing both a favorable and an avertive im- 
pulse. Unconscious conflict may lead also to obsessional and compul- 
sive behavior, symptomatic of the individual’s struggle for release 
from the conflict, but not as such intelligible to him, until interpreted 
by the psychoanalyst. And this is the point d’appui of the analytic 
theory and therapy. 

Although the importance of conflict in the evolution of the indi- 
vidual personality is under no circumstances to be denied, it seems 
that only in exceptional cases is the psychoanalytic emphasis on its 
unconscious operation fully justified. Most conflicts, psychoanalysis 
to the contrary notwithstanding, are conscious in all essential particu- 
lars and for that reason another less esoteric portrayal of conflict 
seems more adequate. 

The constant war between impulsive desires (psychoanalysis calls 
them “instincts”) and various inhibitory agencies (chiefly the Super- 
ego) is sometimes solved by sublimation whereby impulses are ex- 

80 F. Alexander, The Medical Value of Psychoanalysis , 1932, p. 79. 



pressed in socially and personally acceptable channels (e.g., work, 
play, or art) without contingent suffering. Such aim-inhibited wishes 
(e.g., nursing in place of maternity, or “latent” in place of “overt” 
homosexuality) play a large part in the socialization of personality. 
Psychoanalysis regards this as an altogether transparent substitution, 
and one that, so long as it is successful, keeps the individual from 
becoming neurotic. 

Sublimation, it claims, is the device above all others that normal 
personalities employ to render their anti-social impulses acceptable. 
If this were the case, sublimation would be the most important genetic 
mechanism in the study of normal personality, but as the next chapter 
will show the maturing of normal personality is a far more complex 
process than simply redirecting the aim of originally unallowable 
wishes . 81 

Another serviceable concept is that of identification, applied when 
one person develops an emotional tie with some other person to such 
an extent that he behaves as if he 'were that person. The characteristics 
of the second individual are reproduced through conscious or un- 
conscious imitation by the first. Identification is thought to be a large 
factor in the development of the child’s personality, but it is also 
by no means uncommon in adults. Parents, for example, may identify 
with their own children quite as much as their children with them. 

31 Just what sublimation means in concrete application is seldom clear. The 
concept is confused in the minds of psychologists and laymen alike. The following 
fourfold analysis may help. (1) As applied to highly specific organic tensions (hun- 
ger, need for oxygen, or physiological sex processes) the concept has absolutely no 
applicability. One cannot sublimate starvation nor a distended sex-gland. (Cf. W. S. 
Taylor, “A Critique of Sublimation in Males: A Study of Forty Superior Single 
Men.” Genet . Psychol . Monog 1933, 13, No. 1.) Local segmental tensions can only 
be relieved in ways specifically suited to them. (2) As applied to the distraction of 
attention from an unwelcome interest-by keeping otherwise occupied-the concept 
is really a misnomer. One reduces anger by leaving the scene of provocation and 
taking up some absorbing occupation; not by sublimation, but through a redirection 
of attention and interest. (3) As applied to the fatiguing out of generalized con- 
current tensions accompanying a specific state of unrest, the doctrine has more 
merit. The diffuse somatic restlessness induced by thirst, sexual desire, and the like 
can often be reduced by irrelevant activity that fatigues the organism as a whole. 
(The specific segmental tension however is not direcdy relieved by such activity.) 
(4) As a still more complex concept, implying that an individual may without serious 
conflict forego some specific gratification, provided that he finds other sources of 
equal satisfaction, sublimation is a useful doctrine. In such instances, the individual 
simply disregards his unfulfilled desires, letting them atrophy, or repressing them 
without disaster, in the interest of an alternative plan of life that satisfies not these 
desires but satisfies him as a whole man. But in such a case the original psycho- 
analytic definition of the term is violated, for the individual is not sublimating the 
original energy at all. He is busy doing something quite different, namely, leading 
a satisfying life in spite of the lack of fulfillment of a certain desire. 

1 86 


It is doubtful whether any really new psychological process is here 
involved; imitation of an intensely emotional order seems to cover 
the situation. 

Identification raises the problem of the parent-image. It is obvious 
that the prime factor in the development of any personality is the 
influence of other personalities. Of all the people who affect this de- 
velopment, in general the parents do so most poignantly. Psychologi- 
cal studies have failed to find any significant tendency for children 
to prefer the parent of the opposite sex (as Freudian theory assumes). 
Rather, both boys and girls as a rule have a greater fondness for their 
mothers, no doubt because of her closer association with them and 
her devotion to their comfort and well-being. A child seems never 
to be wholly indifferent to his parents, especially as he grows older. 
And whether his affective attitude toward them is positive or nega- 
tive, the parent-image affects him enormously; and he never escapes 
from it. If an orphan has no memories of one or of both parents 
he takes the nearest lying substitutes, or else, if necessary, produces 
fanciful parents for his own guidance and satisfaction. 

Now, the parent-image is a very concrete factor in experience, 
just as individual as is the parent himself. A child knows no other 
mother or father than his own. A father may be stern or gentle; the 
child’s image conforms throughout life to the type; a mother may be 
the full-bosomed domestic type of woman or the high-strung, artistic 
or professional sort of mother, but to the child she is mother. Ordi- 
narily the child is unquestioning in his acceptance of the parent as 
the parent is. To be sure, after early childhood he may compare his 
parents with other parents, but even if he finds his own lacking, he 
generally does not want to change. A lad of eight said to his mothei, 
“I don’t want you to change, for you’re my mother.” The influence 
of the parent is then usually positive, which means that standards, 
tastes, and characteristics of the parent are likely to be imitated. The 
effect is as subtle as it is profound. It has been found for instance 
that most parents apply to their children the same standards and prac- 
tices which their own parents set for them. Thus it comes about that 
a household perpetuates its own inner mores for generations. If a 
reaction against the codes and customs of the older generation has 
taken place, there is in effect a negative imitation, a protest, which 
just as certainly shows the potency of the parent-image . 82 

32 Cf. K. V. Francis and E. A. Fillmore, Univ. Iowa Studies in Child Welfare , 
>934. 9, No. 2 . 



Especially in attitudes toward the opposite sex does the image 
(and therefore the parent) play a role. Men often choose wives in 
subtle ways like their mothers, or in rare cases where the mother is the 
object of ambivalent regard, wives strikingly and significantly differ- 
ent in type. Women choose husbands like their fathers in equally 
subtle ways, in so far as they have freedom of choice. Men often re- 
sent in their wives any departure from the mother-image. A man 
whose mother was domestic, a homemaker, is more likely to resent 
professional or undomestic interests in his wife; whereas a man accus- 
tomed to see his own mother in an intellectual role prefers his wife to 
be the same, or at least accepts it unquestionably if she assumes such 
a role. The wife, in turn, is frequently as quick to measure her hus- 
band by standards embodied in her father-image. 

Finally, attention must be directed to the all-important psycho- 
analytic conception of psychosexuality, comprising “all aspects of 
love and pleasure-seeking and their mutual inter-relationships; it em- 
phasizes unconscious wishes for sensual gratification and their con- 
scious de-erotized derivatives, normal and abnormal, as well as wishes 
which culminate in complete and mature heterosexual union.” 32 
The concept— of which this is a typical definition— is very slippery. 
Taken in one way, it covers every life-serving impulse of the human 
being; it is Eros, and Eros is Life (opposed only by the “Death In- 
stincts”). Such a generalized postulate of a dynamic, non-specific 
libido is nothing other than the harmless hypothesis of a basic Will to 
Live (p. 109), not unlike the doctrine of Egoism previously discussed 
in this chapter. 

But such a broad, and psychologically valueless, conception of 
psychosexuality is not, after all, the psychoanalyst’s real concern. In 
theory and in practice he translates it into a doctrine of sexuality in 
its narrower sense. In fact psychoanalysis, especially the Freudian 
variety, succeeds in the almost impossible task of over-e mphasizing 
the role of sexual motivation and interest in the human person. This 
is no small accomplishment, for-in Western culture at least-sexual 
tensions are in fact the most important single factor in the develop- 
ment of most personalities (cf. pp. n6f.), or rather sex would be 
the most important single factor if there were any single factors, 
which there are not. 

Biological motives never operate singly. Sex in normal lives 

33 1 . Hendrick, Facts and Theories of Psychoanalysis, 1934, p. 299. 

1 8 8 


never stands alone, it is tied to all manner of personal images, 
sanctions, tastes, interests, ambitions, codes, and ideals. To put the 
case as crisply as possible: sexuality in its stark biological simplicity 
is segmental in the organism, often insistent but never devoid of 
mental ramifications; in these ramifications it is indeed pervasive, but 
it is no longer mere sexuality; it becomes diffused into the major 
systems of interests and traits which are themselves the fundamental 
structural and functional systems of personality. 

A remarkably illogical procedure seems to be responsible for over- 
emphasis on sex by psychoanalysts (by Freudians especially). What- 
ever form of behavior or thought is ever found in any life, to be asso- 
ciated with sex, they seem to assume to be always connected with sex 
in every life. This procedure produces such absurdities as interpret- 
ing the infant’s bad memory (p. 161) as guilt repression (the justifi- 
cation being that neurotic adults are known sometimes to dissociate 
painful sexual memories of guilt from their own consciousness); or 
the dogma that all individuals normally have erotic attachments to 
the opposite sexed parent (because some neurotics report incestuous 
impulses). Actually what is true is that the extreme lability of sexual 
life makes all manner of associations and all manner of conflicts pos- 
sible; but not every libidinous attachment or conflict of one person 
is to be regarded as a psychosexual peculiarity of all people. The rami- 
fications of sexual interest are broad enough and deep enough in any 
life without the need to exaggerate their place by making the sexual 
history of certain typical neurotics the prototype for personality in 

In any two personalities sexuality never seems to play the same 
role. Its attachments, its significance, and the conduct associated with 
it are among the most individualistic of all the phenomena of mental 
life. In spite of its biologically uniform aspects, in its psychological 
organization it is a remarkably idiosyncratic matter. 

And this is why sex, as such, cannot be regarded as a single 
factor of motivation, nor as the basic element in personality. A life 
is not simply a variation on a uniform pattern of psychosexuality, 
but, on the contrary, the sexuality of a life can be understood oniy 
if it is regarded as one of the variations within the total and complete 
pattern of personality. Excepting in the most infra-personal sense 
there is no such concrete fact of sex; when one speaks of sex-habits 
and sex-adjustments one can only mean personal habits and personal 
adjustments, having partial but not exclusive reference to the seg- 


mental biological functions of sex. Personality, then, is not a system 
of formations within a matrix of sex. 

What is true of sex is true of every other so-called instinct. Hu- 
man motives are highly individual affairs. It misrepresents them to 
say that they are only changes rung upon universal themes. Motives 
are always dynamic formations of minds-in-particular, and they can 
only be understood if the course of their individual transformations 
is known. The following chapter is devoted to proving this important 



The Me, like every other aggregate, changes as it grows. 

—William James 

Somehow in the process of maturing the manifold potentialities and 
dispositions of childhood coalesce into sharper, more distinctive mo- 
tivational systems. Pari passu with their emergence these systems take 
upon themselves effective driving power, operating as mature, auton- 
omous motives quite different in aim and in character from the 
motivational systems of juvenile years, and very different indeed from 
the crude organic tensions of infancy. 

One of the chief characteristics of the mature personality is its 
possession of sophisticated and stable interests and of a characteristic 
and predictable style of conduct. Convictions and habits of expres- 
sion are definitely centered. Evaluations are sure, actions are precise, 
and the goals of the individual life are well defined. 

G. K. Chesterton gives a brief but psychologically significant 
portrait of a thoroughly mature personality, Leo Tolstoy, in whom 
all motivation seems to be centered in one master-sentiment. 

Tolstoy, besides being a magnificent novelist, is one of the very 
few men alive who have a real, solid, and serious view of life. . . . 
He is one of the two or three men in Europe, who have an attitude 
toward things so entirely their own, that we could supply their in- 
evitable view in anything— a silk hat, a Home Rule Bill, an Indian 
poem, or a pound of tobacco. There are three men in existence who 
have such an attitude: Tolstoy, Mr. Bernard Shaw, and my friend 
Mr. Hillaire Belloc. They are all diametrically opposed to each other, 
but they all have this essential resemblance, that given their basis of 
thought, their soil of conviction, their opinions on every earthly sub- 
ject grow there naturally, like flowers in a field. There are certain 
views of certain things that they must take; they do not form opin- 

* A part of this chapter is here reprinted from The Golden Jubilee Volume of 
the American Journal of Psychology, Vol. jo, 1937. 


ions, the opinions form themselves. Take, for instance, in the case of 
Tolstoy, the mere list of miscellaneous objects which I wrote down 
at random above, a silk hat, a Home Rule Bill, an Indian poem, and a 
pound of tobacco. Tolstoy would say: “I believe in the utmost pos- 
sible simplification of life; therefore, this silk hat is a black abortion.” 
He would say: “I believe in the utmost possible simplification of life; 
therefore this Home Rule Bill is a mere peddling compromise; it is no 
good to break up a centralised empire into nations, you must break 
the nation up into individuals.” He would say: “I believe in the utmost 
possible simplification of life; therefore, I am interested in this In dian 
poem, for Eastern ethics, under all their apparent gorgeousness, are 
far simpler and more Tolstoyan than Western.” He would say: “I be- 
lieve in the utmost possible simplification of life; therefore, this pound 
of tobacco is a thing of evil; take it away.” Everything in the world, 
from the Bible to a bootjack, can be, and is, reduced by Tolstoy to 
this great fundamental Tolstoyan principle, the simplification of life . 1 

One must, of course, dismiss as literary exaggeration Chesterton’s 
claim that there are only “two or three men in Europe” so well in- 
tegrated that one could supply their inevitable view in anything. 
Among our own acquaintances we can name several more. In prin- 
ciple, however, if not in statistics, Chesterton is right, for the major- 
ity of personal lives are not nearly so unified as Tolstoy’s; in few 
cases is the Leitmotif so entirely consistent. The difference is one 
of degree. For in nearly all mature personalities master-sentiments 
exist, and, however difficult the task may be, psychologists are 
bound to try to account for them. 


To understand the dynamics of the normal mature personality a 
new and somewhat radical principle of growth must be introduced 
to supplement the more traditional genetic concepts thus far con- 
sidered. For convenience of discussion this new principle may be 
christened the fimctional autonomy of motives ? 

1 From G. K. Chesterton and others, Leo Tolstoy, 1903, pp. 3 f. 

2 The authenticity of this principle has been admitted by many psychological 
writers, but they have neglected thus far to give it a name. Its most familiar state- 
ment to date is the oft-quoted phrase of R. S. Woodworth, “mechanisms may^ be- 
come drives.” Another clear recognition lies in the following quotation from E. C. 
Tolman. But neither Woodworth nor Tolman has adopted a substantive designation 
for the psychological process in question. 

“The whole body of both what the anthropologists find in the way of specific 
culture-patterns and what psychologists find in the way of individual idiosyncrasies 


Now, any type of psychology that treats motives, thereby en- 
deavoring to answer the question as to why men behave as they do, 
is called a dynamic psychology. By its very nature it cannot be 
merely a descriptive psychology, content to depict the what and the 
how of human behavior. The boldness of dynamic psychology in 
striking for causes stands in marked contrast to the timid, “more 
scientific” view that seeks nothing else than the establishment of a 
mathematical function for the relation between some artificially 
simple stimulus and some equally artificial and simple response. If 
the psychology of personality is to be more than a matter of coeffi- 
cients of correlation it must be a dynamic psychology, and seek first 
and foremost a sound and adequate theory of the nature of human 

Unfortunately the type of dynamic psychology almost universally 
held, however sufficient it may seem from the point of view of the 
abstract motives of abstract personalities, fails to provide a founda- 
tion sound enough or flexible enough to bear the weight of any 
single full-bodied personality. The reason is that all prevailing dy- 
namic doctrines refer every mature motive of personality to underly- 
ing original instincts, wishes, or needs, shared by all men. Thus, the 
concert artist’s devotion to his music might be “explained” as an ex- 
tension of his “self-assertive instinct,” of the “need for sentience,” 
or as a symptom of some repressed striving of “the libido.” In 
McDougall’s hormic psychology, for example, it is explicitly stated 
that only the instincts or propensities can be prime movers. Though 
capable of extension (on both the receptive and executive sides), 
they are always few in number, common in all men, and established 
at birth. The enthusiastic collector of bric-a-brac derives his enthusi- 
asm from the parental instinct; so too does the kindly old philan- 
thropist, as well as the mother of a brood. It does not matter how 
different these three interests may seem to be, they derive their 
energy from the same source. The principle is that a very few basic 
motives suffice for explaining the endless varieties of human interests. 
And the psychoanalyst holds the same over-simplified theory. The 
number of human interests that he regards as so many canalizations 
of the one basic sexual instinct is past computation. 

seems to consist for the most part, psychologically speaking, in acquired specifica- 
tions of ultimate goals or in acquired adherences to specific types of means-objects, 
which latter then often set up in their own right. And such specifications and 
settings-up, once established, acquire a strangle hold.” Phil. Science, 1935, 2, p. 37a 


Taking the case of Tolstoy, Adler would find the style of life 
adopted by Tolstoy to be a consequence of his compensatory striv- 
ing for power, for health, or for personal integrity in the face of an 
unfavorable environment. Freud might decide that the “simplifica- 
tion of life” was a mere ritual evolved to escape feelings of guilt de- 
rived from an unhallowed infantile love; or perhaps he would at- 
tribute it to a Death Wish. Rank would see it as a desire to return 
to the peaceful pre-natal life. Kempf might say that it represented a 
sublimational craving, sustained by a tension produced by unfulfilled 
love or by some danger not successfully averted. McDougall might 
attribute it to the combined effects of the propensities for submission 
and comfort. H. A. Murray might say that there was a need for 
submission and inviolacy. And in the language of W. I. Thomas, the 
wish for security or recognition, perhaps both, would be made re- 
sponsible. Any of these writers, to be sure, would admit that the 
original motive had become greatly extended both in the range of 
stimuli which provoke it and in its varieties of expression. But the 
common factor in all these explanations is the reduction of every 
motive, however elaborate and individual, to a limited number of 
basic interests, shared by all men, and presumably innate. 

The authors of this type of dynamic psychology are concerning 
themselves only with mind-in-general. They seek a classification of 
the common and basic motives of men by which to explain the 
normal or neurotic behavior of any individual case. (This is true 
even though they may regard their own list as heuristic or even as 
fictional.) The plan really doesn’t work. The very fact that the lists 
are so different in their composition suggests— what to a naive ob- 
server is plain enough— that motives are almost infinitely varied 
among men, not only in form but in substance. Not four wishes, nor 
eighteen propensities, nor any and all combinations of these, even 
with their extensions and variations, seem adequate to account for 
the endless variety of goals sought by an endless variety of mortals. 
And paradoxically enough, in certain cases the few simplified needs 
or instincts alleged to be the common ground for all motivation, 
turn out to be completely lacking (cf. p. 112). 

Before describing the principle of functional autonomy, its the- 
oretical significance should stand out clearly. The stress in this 
volume is constantly on the ultimate and irreducible uniqueness of 
personality. “But how,” cry all the traditional scientists, including 
the older dynamic psychologists, “how are we ever to have a science 


of unique events? Science must generalize.” Perhaps it must, but 
what the objectors forget is that a general law may be a law that tells 
how uniqueness comes about. It is manifest error to assume that a 
general principle of motivation must involve the postulation of ab- 
stract or general motives. The principle of functional autonomy, here 
described, is general enough to meet the needs of science, but par- 
ticularized enough in its operation to account for the uniqueness of 
personal conduct. 

The dynamic psychology proposed here regards adult motives 
as infinitely varied, and as self-sustaining, contemporary systems, 
growing out of antecedent systems, but functionally independent 
of them. Just as a child gradually repudiates his dependence on his 
parents, develops a will of his own, becomes self-active and self- 
determining, and outlives his parents, so it is with motives. Each mo- 
tive has a definite point of origin which may lie in the hypothetical 
instincts, or, more likely, in the organic tensions and diffuse irrita- 
bility described in Chapter IV. Theoretically all adult purposes can 
be traced back to these seed-forms in infancy. But as the individual 
matures the bond is broken. The tie is historical, not functional. 

Such a theory is obviously opposed to psychoanalysis and to all 
other genetic accounts that assume inflexibility in the root purposes 
and drives of life. (Freud says that the structure of the Id never 
changes.) The theory declines to believe that the energies of adult 
personality are infantile or archaic in nature. Motivation is always 
contemporary. The life of modern Athens is continuous with the 
life of the ancient city, but it in no sense depends upon it for its 
present “go.” The life of a tree is continuous with that of its seed, 
but the seed no longer sustains and nourishes the full grown tree. 
Earlier purposes lead into later purposes, but are abandoned in their 

William James taught a curious doctrine that has been a matter 
for incredulous amusement ever since, the doctrine of the transitori- 
ness of instincts. According to this theory— not so quaint as sometimes 
thought— an instinct appears but once in a lifetime, whereupon it 
promptly disappears through its transformation into habits. If there 
are instincts this is no doubt their fate, for no instinct can retain 
its motivational force unimpaired after it has been absorbed and re- 
cast under the transforming influence of learning. Such is the reason- 
ing of James, and such is the logic of functional autonomy. The 
psychology of personality must be a psychology of post-instinctive 



behavior. If, as in this volume, instincts are dispensed with from the 
beginning, the effect is much the same, for whatever the original 
drives or “irritabilities” of the infant are, they become completely 
transformed in the course of growth into contemporaneous systems 
of motives. 

Woodworth has spoken of the transformation of “mechanisms” 
into “drives.” 3 A mechanism Woodworth defines as any course of 
behavior that brings about an adjustment. A drive is any neural 
process that releases mechanisms especially concerned with con- 
summatory reactions. In the course of learning, many preparatory 
mechanisms must be developed in order to lead to the consummation 
of an original purpose. These mechanisms are the effective cause of 
activity in each succeeding mechanism, furnishing the drive for each 
stage following in the series. Originally all these mechanisms were 
merely instrumental, only links in the long chain of processes involved 
in the achievement of an instinctive purpose; with time and develop- 
ment, with integration and elaboration, many of these mechanisms 
become activated directly, setting up a state of desire and tension for 
activities and objects no longer connected with the original impulse. 
Activities and objects that earlier in the game were means to an end, 
now become ends in themselves . 4 

Although Woodworth’s choice of quasi-neurological terminology 
is not the best, his doctrine, or one like it, is indispensable in account- 
ing for the infinite number of effective motives possible in human 
life, and for their severance from the rudimentary desires of infancy. 
Further discussion of the operation of the principle and a critique 
of Woodworth’s position will be more to the point after a review 
of the evidence in favor of the principle. 

3 R. S. Woodworth, Dynamic Psychology , 1918. Equivalent assertions are those 
of W. Stern concerning the transformation of “phenomotives” into “genomotives” 
(. Allgemeine Psychologic , 1935, p. 569) ; and of E. C. Toiman regarding the “strangle 
hold” that “means-objects” acquire by “setting up in their own right” (ftn. p. 19 0, 

4 “The fundamental drive towards a certain end may be hunger, sex, pugnacity 
or what not, but once the activity is started, the means to the end becomes an ob- 
ject of interest on its own account” (Woodworth, op. tit*, p. 201), “The primal 
forces of hunger, fear, sex, and the rest, continue in force, but do not by any means, 
even with their combinations, account for the sum total of drives activating the 
experienced individual” (Ibid,, p. 104). 

i 9 6 the transformation of motives 


Let us begin in a common sense way. An ex-sailor has a craving 
for the sea, a musician longs to return to his instrument after an en- 
forced absence, a city-dweller yearns for his native hills, and a miser 
continues to amass his useless horde. Now, the sailor may have first 
acquired his love for the sea as an incident in his struggle to earn a 
living. The sea was merely a conditioned stimulus associated with 
satisfaction of his “nutritional craving.” But now the ex-sailor is per- 
haps a wealthy banker; the original motive is destroyed; and yet the 
hunger for the sea persists unabated, even increases in intensity as it 
becomes more remote from the “nutritional segment.” The musician 
may first have been stung by a rebuke or by a slur on his inferior 
performances into mastering his instrument, but now he is safely 
beyond power of these taunts; there is no need to continue, yet he 
loves his instrument more than anything else in the world. Once 
indeed the city dweller may have associated the hills around his 
mountain home with nutritional and erotogenic satisfactions, but these 
satisfactions he finds in his city home, not in the mountains; whence 
then comes all his hill-hunger? The miser perhaps learned his habits 
of thrift in dire necessity, or perhaps his thrift was a symptom of 
sexual perversion (as Freud would claim), and yet the miserliness 
persists, and even becomes stronger with the years, even after the 
necessity or the roots of the neurosis have been relieved. 

Workmanship is a good example of functional autonomy. A good 
workman feels compelled to do clean-cut jobs even though his 
security, or the praise of others, no longer depend upon high 
standards. In fact, in a day of jerry-building his workman-like 
standards may be to his economic disadvantage. Even so he cannot 
do a slipshod job. Workmanship is not an instinct, but so firm is the 
hold it may acquire on a man that it is little wonder Veblen mistook 
it for one. A business man, long since secure economically, works 
himself into ill-health, and sometimes even back into poverty, for the 
sake of carrying on his plans. What was once an instrumental tech- 
nique becomes a master-motive. 

Neither necessity nor reason can make one contented perma- 
nently on a lonely island or on an isolated country farm after one is 
adapted to active, energetic city life. The acquired habits seem suffi- 


cient to urge one to a frenzied existence, even though reason and 
health demand the simpler life. 

The pursuit of literature, the development of good taste in clothes, 
the use of cosmetics, the acquiring of an automobile, strolls in the 
public park, or a winter in Miami, may first serve, let us say, the 
interests of sex. But every one of these instrumental activities may 
become an interest in itself, held for a lifetime, long after the erotic 
motive has been laid away in lavender. People often find that they 
have lost allegiance to their original aims because of their deliberate 
preference for the many ways of achieving them. 

The maternal sentiment offers an excellent final illustration. Many 
young mothers bear their children unwillingly, dismayed at the 
thought of the drudgery of the future. At first they may be indiffer- 
ent to, or even hate, their offspring; the “parental instinct” seems 
wholly lacking. The only motives that hold such a mother to child- 
tending may be fear of what her critical neighbors will say, fear of 
the law, a habit of doing any job well, or perhaps a dim hope that the 
child will provide security for her in her old age. However gross these 
motives, they are sufficient to hold her to her work, until through the 
practice of devotion her burden becomes a joy. As her love for the 
child develops, her earlier practical motives are forgotten. In later 
years not one of these original motives may operate. The child may 
be incompetent, criminal, a disgrace to her, and far from serving as 
a staff for her declining years, he may continue to drain her resources 
and vitality. The neighbors may criticize her for indulging the child, 
the law may exonerate her from allegiance; she certainly feels no 
pride in such a child; yet she sticks to him. The tenacity of the ma- 
ternal sentiment under such adversity is proverbial . 5 

Such examples from everyday experience could be multiplied ad 
infinitum. The evidence, however, appears in sharper outline when it 
is taken from experimental and clinical studies. In each of the follow- 
ing instances some new function emerges as an independently struc- 
tured unit from preceding functions. The activity of these new units 
does not depend upon the continued activity of the units from which 
they developed. 

5 Most mothers, to be sure, give their babies a somewhat warmer welcome from 
the start, but even so, there is little evidence that the maternal instinct is a ready- 
made, full-fledged and invariable possession of all women. Even those who have 
early learned to be fond of babies find that with practice and experience the inter- 
est becomes constantly stronger, demanding no other satisfaction for itself than its 
own autonomous functioning. Some women become so absorbed in being good 
mothers that they neglect being the good wives they were earlier. 

198 the transformation of motives 

1. The Circular Reflex. Everyone has observed the almost endless 
repetition of acts by a child. The good-natured parent who picks up 
a spoon repeatedly thrown down by a baby wearies of this occupa- 
tion long before the infant does. Such repetitive behavior, found like- 
wise in early vocalization (babbling), and in other early forms of 
play, is commonly ascribed to the mechanism of the circular reflex. 6 
It is an elementary instance of functional autonomy; for any situation 
where the consummation of an act provides adequate stimulation for 
the repetition of the same act does not require any backward tracing 
of motives. The act is self-perpetuating until it is inhibited by new 
activities or fatigue. 

2. Conative Perseveration. Many experiments show that incom- 
pleted tasks set up tensions that tend to keep the individual at work 
until they are resolved. No hypothesis of self-assertion, rivalry, or any 
other basic need, is required. The completion of the task itself has be- 
come a quasi-need with dynamic force of its own. It has been shown, 
for example, that interrupted tasks are better remembered than com- 
pleted tasks; 7 that an individual interrupted in a task will, even in the 
face of considerable opposition, return to that task; 8 that even trivial 
tasks undertaken in a casual way become almost haunting in character 
until they are completed. 9 

Conative perseveration of this order is stronger if an empty interval 
of time follows the period of work, showing that left to itself , with- 
out the inhibiting effect of other duties or activities, the motive grows 
stronger and stronger. The experiment of Kendig proves this point, 
as well as that of C. E. Smith. 10 The latter investigator demonstrated 
that there is more success in removing a conditioned fear if the de- 
conditioning process is commenced immediately. After a twenty-four 
hour delay the fear has become set, and is more difficult to eradicate. 
We are reminded here of the sound advice to drivers of automobiles 
or airplanes who have been involved in an accident, that they drive 
again immediately to conquer the shock of the accident, lest the fear 
become set into a permanent phobia. The rule seems to be that unless 
specifically inhibited all emotional shocks, given time to set, tend ta 
take on a compulsive autonomous character. 

3. “ Conditioned Reflexes ” Not Requiring Reinforcement . The pure 

6 As a means of fixating early habits and of providing a foundation for future 
learning, this mechanism has received detailed attention by E. B. Holt, Animal Drive 
and the Learning Process , 1931, esp. chaps, vii and viii. 

7 B. Zeigarnik, Psychol. Forsch., 1927, 9, 1-85. 

8 M. Ovsiankina, Psychol. Forsch., 1928, 6, 302-379. 

9 1 . Kendig, “Studies in Perseveration” (in five parts), /. Psychol 1936, 3, 223- 
264. : ' ■■ - ; :Sf Si® A 

10 C. E. Smith, Change in the Apparent Resistance of the Skin as a Function oj 
Certain Physiological and Psychological Factors (Harvard College Library), 1934. 


conditioned reflex readily dies out unless the secondary stimulus is 
occasionally reinforced by the primary stimulus (cf. p. 153). The 
dog does not continue to salivate whenever it hears a bell unless some- 
times at least an edible offering accompanies the bell. But there are 
innumerable instances in human life where a single association, never 
reinforced, results in the establishment of a life-long dynamic system. 
An experience associated only once with a bereavement, an accident, 
or a battle, may become the center of a permanent phobia or complex, 
not in the least dependent on a recurrence of the original shock. 

4. Counterparts in Animal Behavior . Though the validity of a prin- 
ciple in human psychology never depends upon its having a counter- 
part in animal psychology, still it is of interest to find functional 
autonomy in the lower organisms. For example, rats, who will first 
learn a certain habit only under the incentive of some specific tension, 
as hunger, will, after learning, often perform the habit even when fed 
to repletion. 11 

Another experiment shows that rats trained to follow a long and 
difficult path, will for a time persist in using this path, even though a 
short easy path to the goal is offered and even after the easier path 
has been learned. 12 Among rats as among human beings, old and use- 
less habits have considerable power in their own right. 

Olson studied the persistence of artificially induced scratching 
habits in rats. Collodion applied to the ears of the animal set up re- 
moving and cleaning movements. Four days later the application was 
repeated. From that time on the animals showed significantly greater 
number of cleaning movements than control animals. A month after 
the beginning of the experiment when the ears of the rats as studied 
by the microscope showed no further trace of irritation, the number 
of movements was still very great. Whether the induced habit spasm 
was permanently retained the experimenter does not say. 13 

5. Rhythm . A rat whose activity bears a definite relation to his 
habits of feeding (being greatest just preceding a period of feeding 
and midway between two such periods) will, even when starved, dis- 
play the same periodicity and activity. The acquired rhythm persists 
without dependence on the original periodic stimulation of feeding, 14 

Even a mollusc whose habits of burrowing in the sand and reappear- 
ing depend upon the movements of the tide, will, when removed from 

11 J. D. Dodgson, Psychobiology , 1917, 1, 231-276. This work has already been 
interpreted by K. S. Lashley as favoring Woodworth’s dynamic theory as opposed 
to Freud’s (Psychol, Rev., 1924, 31, 192-202). 

12 H. C. Gilhouscn, /. Comp. Psychol., 1933, 16, 1-23. 

is w. C. Olson, The Measurement of Nervous Habits in Normal Children, 1929 
pp. 62-65. 

14 c. P. Richter, Comp . Psycho!. Monog. f 1922, 1, No, a. 



the beach to the laboratory, continue for several days in the same 
rhythm without the tide. Likewise certain animals, with nocturnal 
rhythms advantageous in avoiding enemies, obtaining food, or pre- 
venting excessive evaporation from the body, may exhibit such 
rhythms even when kept in a laboratory with constant conditions of 
illumination, humidity, and temperature. 15 

There are likewise instances where acquired rhythms in human life 
have taken on a dynamic character. Compulsive neurotics enter upon 
fugues or debauches, apparently not because of specific stimulation, 
but because “the time has come.” A dipsomaniac in confinement and 
deprived for months of his alcohol describes the fierceness of the re- 
current appetite (obviously acquired). 

Those craving paroxysms occur at regular intervals, three weeks 
apart, lasting for several days. They are not weak, namby-pamby 
things for scoffers to laugh at. If not assuaged with liquor they be- 
come spells of physical and mental illness. My mouth drools saliva, 
my stomach and intestines seem cramped, and I become bilious, 
nauseated, and in a shaky nervous funk. 16 

In such states of drug addiction, as likewise in states of hunger, lust, 
fatigue, there is to be sure a physical craving, but the rhythms of the 
craving are partially acquired, and are always accentuated by the 
mental habits associated with it. For instance, eating in our civilized 
way of life takes place not because physical hunger naturally occurs 
three times a day, but because of habitual rhythms of expectancy. The 
habit of smoking is much more than a matter of craving for the spe- 
cific narcotic effects of tobacco; it is a craving for the motor ritual 
and periodic distraction as well 

6. Neuroses . Why are acquired tics, stammering, sexual perver- 
sions, phobias, and anxiety so stubborn and so often incurable? Even 
psychoanalysis, with its deepest of depth-probing seldom succeeds In 
effecting complete cures in such cases, even though the patient may 
feel relieved or at least reconciled to his difficulties after treatment. 
The reason seems to be that what are usually called “symptoms” are 
in reality something more. They have set themselves up in their own 
right as independent systems of motivation. Merely disclosing their 
roots does not change their independent activity. 17 

16 S. C. Crawford, Quar. Rev . Biol., 1934, 9, 201-2x4. 

16 Inmate Ward Eight, Beyond the Door of Delusion , Macmillan, 1932, p. 2S1, 

17 The case of W. E. Leonard, The Locomotive God, 1927, is instructive in 
this regard. An intense phobia was not relieved by tracing its history backward to 
the start of life. Even though he could explain why he was once frightened for a 
very good reason (by a locomotive), the author is quite unable to explain why now 
he is frightened for no particular reason . Such neuroses, and psychotic delusional 
systems as well, often acquire a “strangle hold,” and the task of dislodging them is 
usuallv mote than therapeutic skill is equal to. 


7. The Relation Between Ability and Interest . Psychometric studies 
have shown that the relation between ability and interest is always 
positive, often markedly so. A person likes to do what he can do well. 
Over and over again it has been demonstrated that the skill learned 
for some external reason, turns into an interest, and is self-propelling, 
even though the original reason for pursuing it has been lost. A stu- 
dent who at first undertakes a field of study in college because it is 
prescribed, because it pleases his parents, or because it comes at a con- 
venient hour, often ends by finding himself absorbed, perhaps for life, 
in the subject itself. He is not happy without it. The original motives 
are entirely lost. What was a means to an end has become an end in 

And there is the case of genius. A skill takes possession of the man. 
No primitive motivation is needed to account for his persistent, ab- 
sorbed activity. It just is the alpha and omega of life to him. It is im- 
possible to think of Pasteur’s concern for health, food, sleep or fam- 
ily, as the root of his devotion to his work. For long periods of time 
he was oblivious of them all, losing himself in the white heat of re- 
search for which he had been trained and in which he had acquired a 
compelling and absorbing interest. 

A much more modest instance is the finding of industrial research 
that when special incentives are offered and work speeded up as a 
consequence, and then these special incentives removed, the work 
continues at the speeded rate. The habit of working at a faster tempo 
persists without external support. 

8. Sentiments vs. Instincts . Every time an alleged instinct can by 
rigid analysis be demonstrated not to be innate but acquired, there is 
in this demonstration evidence for functional autonomy. It is true 
enough that maternal conduct, gregariousness, curiosity, workman- 
ship, and the like, have the tenacity and compelling power that in- 
stincts are supposed to have. If they are not instincts, then they must 
be autonomous sentiments with as much dynamic character as has 
been attributed to instincts. It is not necessary here to review all the 
arguments in favor of regarding such alleged instincts as acquired 
sentiments; the problem was discussed in Chapter IV. 

9. The Dynamic Character of Personal Values . When an interest- 
system has once been formed it not only creates a tensional condition 
that may be readily aroused, leading to overt conduct in some way 
satisfying to the interest, but it also acts as a silent agent for selecting 
and directing any behavior related to it. Take the case of people with 
strongly marked esthetic interests. Experiments with the word- 
association test have shown that such people respond more quickly to 
stimulus words connected with this interest than to words relating to 



interests they lack . 13 Likewise, in scanning a newspaper they will ob- 
serve and remember more items pertaining to art; they also take a 
greater interest in clothes than do non-esthetic people; and when they 
are asked to rate the virtues of others, they place esthetic qualities 
high. In short the existence of a well-established acquired interest 
exerts a directive and determining effect on conduct just as is to be 
expected of any dynamic system. The evidence can be duplicated for 
many interests other than the esthetic . 19 


Objections to the principle of autonomy may be expected from 
two sides. Behavioristically inclined psychologists will continue to 
prefer their conception of organic drive with its capacity for mani- 
fold conditioning by ever receding stimuli. Whereas instinct psy- 
chology of the traditional order will be unable to accept a pluralistic 
principle that seems to leave purpose so largely at the mercy of 

The behaviorist is well satisfied with motivation in terms of 
organic drive and conditioning because he feels that he somehow has 
secure anchorage in the neural structure. (For some strange reason, 
the closer he approaches nervous tissue the happier the behaviorist 
is.) But the truth of the matter is that the neural physiology of 
organic drive and conditioning is no better established, and no easier 
to imagine, than is the neural physiology of the type of complex 
autonomous units of motivation we have described. 

Two behavioristic principles will be said to account adequately 
for the instances of functional autonomy previously cited, viz., the 
circular reflex and cross-conditioning. The former concept, accept- 
able enough when applied to infant behavior, merely says that the 
more activity a muscle engages in, the more activity of the same sort 
does it engender through a self-sustaining circuit . 20 This is, to be sure, 
a clear instance of autonomy, albeit on a primitive level, over-simpli- 
fied so far as adult conduct is concerned. The doctrine of cross- 
conditioning refers to subtle recession of stimuli in the process of 
conditioning, and to the intricate possibility of cross-connections in 
conditioning. For instance, such ubiquitous external stimuli as hu- 
midity, daylight, gravitation, may feed collaterally into open channels 

18 H. Cantril, “General and Specific Attitudes,” Psychol. Monog., 1932, No. 192. 
29 H. Cantril and G. W. Allport, /. Abnorm. Soc. Psychol., 1933, 28, 239-273. 
20 E. B. Holt, op. cit., p. 38. 


of activity, arousing mysteriously and unexpectedly a form of con* 
duct to which they have unconsciously been conditioned. For exam- 
ple, the angler whose fishing expeditions have been accompanied by 
sun, wind, or a balmy June day, may feel a desire to go fishing 
whenever the barometer, the thermometer, or the calendar in his city 
home tells him that these conditions prevail. 21 Innumerable such 
crossed stimuli are said to account for the arousal of earlier patterns 
of activity. 

Such a theory is highly mechanistic. It inherits, first of all, the 
difficulties resident in the principle of conditioning whenever it is 
applied to human behavior in general (cf. pp. 151-153). Further, 
though the reflex circle and cross-conditioning may in fact exist, they 
are really rather trivial principles. They leave the formation of inter- 
est and its occasional arousal almost entirely to chance factors of 
stimulation. They give no picture at all of the spontaneous and vari- 
able aspects of traits, interests, or sentiments. These dispositions are 
regarded as purely reactive in nature; the stimulus is all-important. 
The truth is that dispositions sort out stimuli congenial to them, and 
this activity does not in the least resemble the rigidity of reflex re- 
sponse. 22 

A variant on the doctrine of cross-conditioning is the principle 
of redintegration , 23 This concept admits the existence of highly inte- 
grated dispositions of a neuropsychic order. These dispositions can be 
aroused as a whole by any stimulus previously associated with their 
functioning. In this theory likewise the disposition is regarded as a 
rather passive affair, waiting for reactivation by some portion of the 
original stimulus. Here again the variability of the disposition and its 
urge-like quality are not accounted for. The stimulus is thought 
merely to reinstate a complex determining tendency. Nothing is said 
about how the stimuli themselves are selected, why a motive once 
aroused becomes insistent, surmounting obstacles, skillfully subordi- 
nating conflicting impulses, and inhibiting irrelevant trains of thought. 

In certain respects the principle of autonomy stands midway be- 
tween the behavioristic view and the thoroughgoing purposive psy- 

21 Ibid,, p. 224. 

22 The basic fact that complex “higher” centers have the power of inhibiting, 
selecting, and initiating the activity of simpler segmental responses is a fact too well 
established to need elaboration here. It constitutes the very foundation of the psycho- 
physiological theories advanced by Sherrington, Herrick, Dodge, Kohler, Troland* 
and many others. 

23 Cf. H. L. Hollingworth, Psychology of the Functional Neuroses , 1920. 

204 THE transformation of motives 

chology of the hormic order. It agrees with the former in emphasiz- 
ing the acquisition of motives, in avoiding an a priori and unchanging 
set of original urges, and in recognizing (as limited principles) the 
operation of the circular response and cross-conditioning. It agrees 
with the hormic psychologist, however, in finding that striving-from- 
within is a far more essential characteristic of motive than stimulation- 
from-without. It agrees likewise in distrusting the emphasis upon 
stomach contractions and other “excess and deficit stimuli” as 
“causes” of mature behavior. Such segmental sources of energy even 
though conditioned cannot possibly account for the “go” of conduct. 
But functional autonomy does not rely as does hormic theory upon 
modified instinct, which after all is as archaic a principle as the con- 
ditioning of autonomic segmental tensions, but upon the capacity of 
human beings to replenish their energy through a plurality of con- 
stantly changing systems of dynamic dispositions. 

The hormic psychologist, however, will not accept the autonomy 
of new motivational systems. If mechanisms can turn into drives, he 
asks, why is it that habits and skills as they become exercised to the 
point of perfection do not acquire an ever increasing driving force ? 24 
The mechanisms of walking, speaking, or dressing, cannot be said 
to furnish their own motive-power. One walks, speaks, or dresses in 
order to satisfy a motive entirely external to these learned skills . 25 
The criticism is sufficiently cogent to call into question Woodworth’s 
form of stating the principle, viz., “Mechanisms may become drives.” 
It is not an adequate statement of the case. 

Looking at the issue more closely it seems to be neither the per- 
fected talent nor the automatic habit that has driving power, but the 
imperfect talent and the habit-in-the-making. The child who is just 
learning to speak, to walk, or to dress, is, in fact, likely to engage in 
these activities for their own sake, precisely as does the adult who 
has an unfinished task in hand. He remembers it, returns to it, and 
suffers a feeling of frustration if he is prevented from engaging in it. 
Motives are always a kind of striving for some form of completion; 
they are unresolved tension, and demand a “closure” to activity under 
way. (Latent motives are dispositions that are easily thrown by a 
stimulus or by a train of associations into this state of active tension.) 

24 W. McDougall, Mind, 1920, N. S,, 29, 277-293, 

25 Though this objection is usually valid, it is not always so, for there are cases 
where the liking for walks, for talking for the sake of talking, or for dressing, play- 
ing games, etc., seem to be self-sustaining motivational systems. 


The active motive subsides when its goal is reached, or in the case 
of a motor skill, when it has become at last automatic. The novice in 
automobile driving has an unquestionable impulse to master the skill. 
Once acquired the ability sinks to the level of an instrumental dis- 
position and is aroused only in the service of some other driving 
(unfulfilled) motive. 

Now, in the case of the permanent interests of personality, the 
situation is the same. A man whose motive is to acquire learning, or 
to perfect his craft, can never be satisfied that he has reached the end 
of his quest, for his problems are never completely solved, his skill 
is never perfect. Lasting interests are recurrent sources of discontent, 
and from their incompleteness they derive their forward impetus. 
Art, science, religion, love, are never perfected. But motor skills are 
often perfected, and beyond that stage they seldom provide their own 
motive power. Only skills in the process of perfecting (mechanisms- 
on-the-make) serve as drives. With this emendation, Woodworth’s 
view is corrected, and McDougall’s objection is met. 28 

If the dynamic psychologist finds in such a pluralistic system a 
displeasing lack of unity, he may, without damage to the principle 
of autonomy, fall back upon the elemental horme. All motives— 
diverse as they ar e—may be regarded as so many channels of the origi- 
nal Will-to-Live. (Such a monistic under-pinning to a theory of 
motivation is preferable to a list of arbitrarily distinguished propensi- 
ties or instincts.) But, as was previously pointed out (p. 109), the 
Will-to-Live, however acceptable it may be in the underlying meta- 
physics of personality, does not itself aid in the task of psychological 

Only such a principle as that under discussion can provide a flexi- 
ble enough account of the plurality of motives and their countless ex- 
pressions in human life. Its specific advantages stand out in the follow- 
ing summary: 

1. It clears the way for a completely dynamic psychology of 
traits, attitudes, interests, and sentiments, which can now be regarded 
as the ultimate and true dispositions of the mature personality. 

2. It avoids the absurdity of regarding the energy of life now, in 
the present, as somehow consisting of early archaic forms (instincts, 
prepotent reflexes, or the never-changing Id). Learning brings new 

2*5 This theory embraces very easily the work of K. Lewin and his associates 
upon the nature of “quasi-needs.” The urgency of these needs is greatest just before 
a goal is reached, after which time die motive subsides completely. 

206 the transformation of motives 

systems of interests into existence just as it does new abilities and 
skills. At each stage of development these interests are always con- 
temporary; whatever drives, drives now. 

3. It dethrones the stimulus. A motive is no longer regarded as a 
mechanical reflex or as a matter of redintegration, depending entirely 
upon the capricious operation of a conditioned stimulus. In a very 
real sense dispositions select the stimuli to which they respond, even 
though some stimulus is required for their arousal. 

4. It readily admits the validity of all other established principles 
of growth. Functional autonomy utilizes the products of differentia- 
tion, integration, maturation, exercise, imitation, suggestion, condi- 
tioning, trauma, and all other processes of development; and allows, 
as they do not, considered by themselves, for their structuration into 
significant motivational patterns. 

5. It places in proper perspective the problems of the origin of 
conduct by removing the fetish of the genetic method. Not that the 
historical view of behavior is unimportant for a complete understand- 
ing of personality, but so far as motives are concerned the cross- 
sectional dynamic analysis is more significant. Motives being always 
contemporary should be studied in their present structure. Failure 
to do so is probably the chief reason why psychoanalysis meets so 
many defeats, as do all other therapeutic schemes relying too ex- 
clusively upon uncovering the motives of early childhood. 

6. It accounts for the force of delusions, shell shock, phobias, and 
all manner of compulsive and maladaptive behavior. One would ex- 
pect such unrealistic modes of adjustment to be given up as soon as 
they are shown to be poor ways of confronting the environment. In- 
sight and the law of effect should both remove them. But too often 
they have acquired a strangle hold in their own right. 

7. At last we can account adequately for socialized and civilized 
behavior. The principle supplies the correction necessary to the 
faulty logic of bellum omnium contra omnes. Starting life as a com- 
pletely selfish being, the child would indeed remain entirely wolfish 
and piggish throughout his days unless genuine transformations of 
motives took place. Motives being completely alterable, the dogma 
of Egoism turns out to be a callow and superficial philosophy of be- 
havior, or else a useless redundancy. 

8 . It explains likewise why a person often becomes what at first 
he merely pretends to be— the smiling professional hostess who grows 
fond of her once irksome role and is unhappy when deprived of it; 


the man who for so long has counterfeited the appearance of self- 
confidence and optimism that he is always driven to assume it; the 
prisoner who comes to love his shackles. Such personae, as Jung ob- 
serves, are often transformed into the real self. The mask becomes 
the anima 27 

9. The drive behind genius is explained. Gifted people demand 
the exercise of their talents, even when no other reward lies ahead. 
In lesser degree the various hobbies, the artistic, or the intellectual 
interests of any person show the same significant autonomy. 

10. In brief, the principle of functional autonomy is a declara- 
tion of independence for the psychology of personality. Though in 
itself a general law, at the same time it helps to account, not for the 
abstract motivation of an impersonal and purely hypothetical mind- 
in-general as do other dynamic principles, but for the concrete, viable 
motives of any one mind-in-particular. 


A special instance of functional autonomy is the effect of abrupt 
shocks on the developing personality. Ordinarily the process of 
growth is gradual; it is like the slow reaching of tentacles in many 
directions, some of the movements being halted when they are found 
maladaptive, and others continued in directions found to make for 
successful survival. All of the processes of growth thus far described— 
with the possible exception of maturation— manifest themselves as a 
rule by this gradual operation. Yet, sometimes, this operation is 
abruptly altered. An entirely new direction is given to the person’s 
aims, outlooks, and style of life. Growth at this moment ceases to 
be gradual and becomes, for the time being, saltatory. 28 

27 C. J. Jung, Psychological Types, 1924, p. 593. 

28 One hundred years ago Charles Fourier, the self-styled “super-omnigyne” 
whose novel theories of personality were described in Chapter III, offered the fol- 
lowing rhythmic scheme for representing the alternating phases of gradual and 
saltatory development. Over-simplified though it is, the list has some suggestive 

f initial crisis— birth 
73 J 1st phase of growth— childhood 
S J ascending crisis— puberty 
f 2nd phase of growth-adolescence 

so /' climax of life— virility 

;§ \ 3rd phase of movement-maturity 
g < descending crisis— sterility 

^ J 4th phase of movement— decline 

Q k final crisis-death 



No one can tell what catastrophic events an individual will en- 
i counter in the course of his life, or what their impression will be 

) upon him. Some life-histories seem pivoted upon one decisive event, 

1 the vision of St. Paul, for example, the illness of St. Francis, the 

j Italian Journey of Goethe, or Nietzsche’s infection by a prostitute, 

j Yet similar experiences in the lives of others have no such radical 

| effects. It is small wonder that William James wrote: “However 

| closely psychical changes may conform to law, it is safe to say that 

| individual histories and biographies will never be written in advance 

J no matter how ‘evolved’ psychology may become.” 29 

| In each of the periods of life (cf. p. 1 3 1 ) there fall certain charac- 

' teristic emergencies that the individual must meet in his struggles for 

adjustment. Though these emergencies do not always result in 
psychic traumas, they do sometimes serve to halt abruptly one course 
1! of development and to start a distinctly new pattern of habits and 

' traits. 

i In infancy, there is first of all the possibility of a birth-trauma, 

j though in view of the immature condition of the nervous system at 

j the commencement of life it is difficult to see how one can attach as 

much weight to this possibility as certain psychoanalysts do. 30 Be 
that as it may, there are undeniably other traumas in infancy that 
may leave permanent effects on personality— accidents, for example, 
or illnesses (e.g., Jacksonian epilepsy or encephalitis). At any time in 
life, for that matter, the traumatic effects of accident and illness may 
alter the preceding direction of development and substitute an alto- 
j gether different one to accord with the changed physical condition. 

The pre-school child does not as a rule encounter crises outside 
j the home-circle, but within the home many critical experiences 

• may occur to redirect the whole course of development: the arrival 

of a new baby with the consequent feelings of jealousy, early experi- 
ences of shame, of bereavement, or perhaps adoption into a foster- 
j j family. 

j I When at about the age of six the child leaves the shelter of his 

j home for the harder environment of the school and playground, ex- 

i periences of failure, ostracism, and ridicule await him, and these may 

i provoke quite suddenly new forms of adjustment or else accentuate 

? previously insignificant traits. The sensitive child may grow definitely 

I .. morbid. Inferiority complexes may be created over night, affecting 

! ® Principles of Psychology, 1890, II, 576 ftn. 
j 80 Cf. O. Rank, The Trauma of Birth, trans. 1929. 



profoundly the subsequent course of life. In the development of most 
boys there is a critical “sissy hurdle” that must be met. Perhaps a 
fist-fight, a foot race, a “grown-up” haircut, or some act of daring 
does the trick. If so a “normal boy’s” life lies ahead for him. But 
perhaps the hurdle is not successfully passed, and as a result the plan 
of life is radically altered; new compensations and new ideals de- 

Soon come the demands of the gang, relations to the opposite sex, 
religious interests. Experiences of success and failure, of remorse and 
guilt, of conversion, or puppy love, may be of supreme importance. 
Also in adolescence comes the frequently traumatic experience of 
leaving home, of being “psychologically weaned” from the parents. 
There are also new worlds to conquer, college examinations to be 
met, a living to be earned, where traumatic experiences of success or 
failure can occur. 31 

In adult years shocks due to business failure, to illness, to religious 
conversion, 32 to the death of loved persons, to the “descending crisis,” 
all may make swift and profound alterations. Yet, as a rule, person- 
ality after the age of thirty is much less subject to sudden upheavals 
than prior to that age. Critical and abrupt changes are never so 
numerous as in adolescence. 

Biography is full of illustrations, some of the most interesting of 
which concern the sudden intrusion of an idea into a preceding stag- 
nant condition of thought. Gibbon dates the first occurrence of his 
ambition to write of the decline and fall of the Roman Empire as 
the evening of the 15th of October, 1764, as he sat musing among the 
ruins of the Capitol. 33 Alice James describes the lasting importance 
of an idea, conveyed to her by her literary brother when she was 
only eight. 

31 The crises of adolescence are interestingly dealt with by L, S, Hollrngworth, 
The Psychology of the Adolescent , 1928. But in combating the '‘widespread myth 
that every child is a changeling, who at puberty comes forth as a different per- 
sonality,”' the author seems to risk the opposite error of underestimating the fre- 
quency with which radical alterations of personality do occur in the period of 
Sturm und Drang . 

32 Conversions, though much more numerous in adolescence, as a rule leave a 
more marked impress upon personality if they come later in life. A well known 
example is the case of Count Tolstoy' A profound religious experience at the age 
of fifty resulted in a radical shift of his whole plan of life and was responsible for 
the development of the new and firmly knit master-motive described by Chesterton 
(pp. 190 f.). 

83 G. B. Hill (editor), The Memoirs of the Life of Edward Gibbon , 1900, n 
167. ' '"r 


I remember so distinctly the first time I was conscious of a purely 
intellectual process. . . . We were turned into the garden to play. 

. . . Harry suddenly exclaimed: “This might certainly be called 
pleasure under difficulties!” The stir of my whole being in response 
to the exquisite, original form of his remark almost makes my heart 
beat now with the sisterly pride which was then awakened, and it 
came to me in a flash— the higher nature of this appeal to the mind, 
as compared to the rudimentary solicitations which usually produced 
my childish explosions of laughter, and I can also feel distinctly the 
sense of self-satisfaction in that I could not only perceive but appre- 
ciate this subtlety, as if I had acquired a new sense, a sense whereby 
to measure intellectual things, wit as distinguished from giggling, for 
example . 34 

Sometimes casual remarks made by other people, especially people 
toward whom because of prestige one is suggestible, have lasting in- 
fluence. A certain freshman in college, coming from an uncultured 
home with exclusively economic aspirations, had a marked super- 
cilious attitude toward all intellectual activities. Badly adjusted to 
college, with poor habits of study, and as nearly illiterate as a college 
youth can be, he had special difficulties with his course in English 
composition. One day when his weary instructor was giving him a 
periodic berating, the boy countered with one of his customary de- 
fenses, “I don’t like English, and I never did like it, and never shall.” 
The instructor, bored but still didactic, remarked, “It isn’t English 
I’m talking about at the moment, it’s your life.” The effect was 
wholly unpredictable. The thrust struck home. The lad not only re- 
formed his precarious ways, but became devoted to the subject, ob- 
tained a high grade, made Phi Beta Kappa, and eventually became a 
teacher of English! 

An older woman traces her life-long devotion to poetry largely 
to an episode in her Virgil class in high school. She was performing 
a routine translation, difficult and dreary. The teacher trying for the 
thousandth time to give significance to the monotonous task, asked 
how the passage would be expressed in the Bible. Happily the girl 
caught the allusion, and spontaneously revised her translation, “In- 
cline thine ear unto my supplication.” Her artistic success on this 
occasion was a traumatic experience, a dawn of poetic beauty, to 
which she was ever after devoted. 

These “chambers of maiden thought,” as Keats has called them, 

“ Alice James, Her Brothers— Her Journal, Dodd, Mead, 1934, p. 1 66. 


are usually entered-if they are entered at all— as the personality ap- 
proaches maturity. The experience may be due to books read, or to 
sermons and lectures heard, to the influence of friends, parents, or 
teachers; it may lead further into the chambers of abstract thought 
and scientific research. It is, of course, not always traumatic in nature, 
but may be a gradual growth, or turn out to be merely abortive. 
However entered, if it is entered at all, the world of ideas is a factor 
that shapes the more complex reaches of personality, and not infre- 
quently it is the most important factor of all. 

Quick turns do then occur in personality and permanently hold 
the stage; the next step is to explain them. Probably the answer must 
be given in terms of the adaptive process . 35 A crisis is brought about 
by an exceedingly intense emotional stimulus. Why it is intense for 
the particular individual depends almost wholly upon the present con- 
dition of susceptibility of that individual (a confluence of his own 
temperament and previous relevant experiences). The crisis throws 
into new relief the factors in a situation that was already, perhaps 
unconsciously, of some importance to him. But though the shift 
occurs upon some pre-existent familiar ground the old familiar pre- 
formed habits and traits no longer meet the need. The crisis imperi- 
ously demands new and more concordant systems. It is so urgent that 
it cannot be set aside, and since it cannot be admitted into the oldei 
setting, a new setting must be swiftly prepared. Sometimes as in the 
case of a religious or moral conversion, the majority of the previous 
habits and attitudes may have to be radically altered. As a result the 
“new” personality seems utterly different from the old. 

It is the nature of traumatic experiences that they are always 
specific, that is to say, they can be dated and defined, but their 
effects are always generalized, spreading into many, or sometimes all, 

35 One type of psychological writing, the dialectical , finds no difficulty what- 
ever in accounting for the phenomenon of sudden reorientation. It holds that the 
entire course of life is a matter of conflict, personality being constantly assailed by 
tendencies to act and by negations of these tendencies, with its goal always the 
synthesis of these conflicting impulses if possible. Sometimes the synthesis 'is gradual, 
but it often comes as a sudden change. Especially when one of the conflicting fac- 
tors is suddenly introduced it is norma! to expect a Katastrophenreaktion (Kiinkel), 
whereby a novel convergence of the old and the new takes place. Good examples 
of the ^dialectical method applied to personality are found in F. Klinkel, Vitale 
Uialektik , 1929, K. K Kornilov, “Psychology in the Light of Dialectical Material- 
ism,” in Psychologies of 1930, chap, xiii, W. Stern, Allgememe Psychologie auf per- 
sonalistischer Grundlage , 1935. Dialectic runs the danger of overestimating the preva- 
lence of sudden, and underestimating the occurrence of gradual, change. 


of the recesses of personality. The newly created interests are 
promptly charged with dynamic power, displacing older formations, 
and henceforth serving as functionally autonomous systems, guiding 
the further development of the personality until they in turn are 
gradually or suddenly transformed. 



Nothing requires a rarer intellectual heroism than willing- 
ness to see one’s equation written out. 


TT* he distinctive richness and congruence of a fully mature person- 
ality are not easy to describe. There are as many ways of growing up 
as there are individuals who grow, and in each case the end-product 
is unique. But if general criteria are sought whereby to distinguish a 
fully developed personality from one that is still unripe, there are 
three differentiating characteristics that seem both universal and in- 

In the first place, the developed person is one who has a variety 
of autonomous interests: that is, he can lose himself in work, in con- 
templation, in recreation, and in loyalty to others. He participates 
with warmth and vigor in whatever pursuits have for him acquired 
value. Egocentricity is not the mark of a mature personality. Contrast 
the garrulous Bohemian, egotistical, self-pitying, and prating of self- 
expression, with the man of confident dignity who has identified him- 
self with a cause that has won his devotion. Paradoxically, “self- 
expression” requires the capacity to lose oneself in the pursuit of ob- 
jectives, not primarily referred to the self. Unless directed outward 
toward socialized and culturally compatible ends, unless absorbed in 
causes and goals that outshine self-seeking and vanity, any life seems 
dwarfed and immature. 

Whenever a definite objective orientation has been attained, pleas- 
ures and pains of the moment, setbacks and defeats, and the impulse 
for self-justification fade into the background, so that they do not 
obscure the chosen goals. These goals represent an extension of the 
self which may be said to be the first requirement for maturity in 
personality. . 

The second requirement is a curiously subtle factor comple- 



of the recesses of personality. The newly created interests are 
promptly charged with dynamic power, displacing older formations, 
and henceforth serving as functionally autonomous systems, guiding 
the further development of the personality until they in turn are 
gradually or suddenly transformed. 



Nothing requires a rarer intellectual heroism than willing- 
ness to see one’s equation written out. 


TT he distinctive richness and congruence of a fully mature person- 
ality are not easy to describe. There are as many ways of growing up 
as there are individuals who grow, and in each case the end-product 
is unique. But if general criteria are sought whereby to distinguish a 
fully developed personality from one that is still unripe, there are 
three differentiating characteristics that seem both universal and in- 

In the first place, the developed person is one who has a variety 
of autonomous interests: that is, he can lose himself in work, in con- 
templation, in recreation, and in loyalty to others. He participates 
with warmth and vigor in whatever pursuits have for him acquired 
value. Egocentricity is not the mark of a mature personality. Contrast 
the garrulous Bohemian, egotistical, self-pitying, and prating of self- 
expression, with the man of confident dignity who has identified him- 
self with a cause that has won his devotion. Paradoxically, “self- 
expression” requires the capacity to lose oneself in the pursuit of ob- 
jectives, not primarily referred to the self. Unless directed outward 
toward socialized and culturally compatible ends, unless absorbed in 
causes and goals that outshine self-seeking and vanity, any life seems 
dwarfed and immature. 

Whenever a definite objective orientation has been attained, pleas- 
ures and pains of the moment, setbacks and defeats, and the impulse 
for self-justification fade into the background, so that they do not 
obscure the chosen goals. These goals represent an extension of the 
self which may be said to be the first requirement for maturity in 
personality. ■' y SS-lT 

The second requirement is a curiously subtle factor comple- 


2 14 

meriting the first. We may call it self -objectification, that peculiar 
detachment of the mature person when he surveys his own preten- 
sions in relation to his abilities, his present objectives in relation to 
possible objectives for himself, his own equipment in comparison with 
the equipment of others, and his opinion of himself in relation to the 
opinion others hold of him. This capacity for self-objectification is 
insight, and it is bound in subtle ways with the sense of humor, which 
as no one will deny, is, in one form or another, an almost invariable 
possession of a cultivated and mature personality. 

Since there is an obvious antithesis between the capacity for losing 
oneself in vigorous participation and the capacity for standing off, 
contemplating oneself, perhaps with amusement, a third, integrative, 
factor is required in the mature personality, namely, a unifying 
philosophy of life. Such a philosophy is not necessarily articulate, 
at least not always articulate in words. The preacher, by virtue of his 
training, is usually more articulate than the busy country doctor, the 
poet more so than the engineer, but any of these personalities, if 
actually mature, participates and reflects, lives and laughs, according 
to some embracing philosophy of life developed to his own satisfac- 
tion and representing to himself his place in the scheme of things. 

These are the three conditions for the optimum development of 
personality; each needs more detailed consideration. But before 
settling to this discussion it is necessary to examine the competence of 
psychology to deal with such complex mental conditions as are here 

When the academic psychologist attempts to account for such 
intricate formations as these he faces a dilemma. He himself lives and 
works primarily in professional circles, among persons trained to use 
their minds and abilities. It is too easy to assume that such men are 
representative of the majority of personalities, and to overlook those 
with gross limitations, occurring more frequently in an unselected 
population. He might easily romanticize the situation by compound- 
ing a representative personality on too high and subtle an emotional 
and mental level. He must not, in his interest in mature personalities, 
be unrealistic and forget the restrictions on the development of per- 
sonality resulting from low intelligence, uncontrolled emotion, in- 
fantilism, regression, dissociation, stereotypes, autism, suggestibility, 
and many other entirely human, but none the less abortifacient con- 

On the other hand, even though the number of completely mature 


personalities tested by these three criteria, may be few, it is still neces- 
sary to account adequately for such as do exist, and for the multitude 
of others well along on the road to maturity. This necessity leads us 
to the second horn of the dilemma, the one on which most psycholo- 
gists, without realizing it, are already impaled. 

They are impaled because they apply the (as yet) crude tools of 
psychology to material too delicate to be cut and shaped with their 
aid. For instance, methods and concepts well enough designed to ex- 
plain the automatic reactions of decorticated cats or simple skin 
reflexes, are often superimposed upon the vast pattern of mature per- 
sonality, and declared to overspread it exactly. The results are ridicu- 
lous, and are responsible for the view of so many educated people 
that psychology is a sappy science. 

Even the psychologist who honestly desires not to underestimate 
the complexities of personality finds himself limited by the crudity 
of the tools within his professional store. As a result he puts the entire 
strain of investigation upon a few inadequate implements where it 
would be better to forge new ones. 

That the available store of concepts and methods is actually insuffi- 
cient can be demonstrated by a cursory review of the major limita- 
tions of the various branches of psychology concerned with person- 
ality. Here it is not a question of the validity of the concepts nor of 
their suitability for special problems, but rather of their adequacy in 
dealing with the subtle characteristics of genuinely mature personali- 

Physiological psychology manifestly fails to specify neural equiva- 
lents for complex personal functions, for such subtle processes, for 
example, as ambition, loyalty, self-criticism, or humor. 

Behaviorism of the classical order provides at best a blue-print of 
non-related excursions and meaningless movements of a mindless or- 
ganism. Its concepts are better adapted to segmental responses than to 
fully integrated patterns. Whether the newer “operational behavior- 
ism” can improve the situation remains for it to demonstrate. 

Structural psychology of the older introspective order, since it is 
not interested in conation, can account for no single dynamic event in 
the entire sphere of personality. 

Functional psychology, like structural, is preoccupied with mind-in- 
general, and though it treats instincts, adaptation, the stream of 
thought, and other vital functions, it does not do so in a personalized, 
concrete way. 

Gestalt psychology, for all-its advantageous emphasis on wholeness,. 

216 the mature personality 

has as yet dealt chiefly with momentary patterns of conduct, and 
tends to neglect the problems of lasting structure. 

Mathematical psychology , with the aid of highly trained and subtle 
minds, produces only a caricature of such minds by holding that they 
can all be reduced to a few basic and common factors, or can be re- 
garded as measurable deviations from one standard pattern. 

Differential psychology , likewise occupied with the distribution of 
single qualities in a population, cannot treat the patterning of these 
differences into individual dynamic formations. 

Freudian psychology never regards an adult as truly adult. 

Dynamic psychology often commits this error of Freudianism, and 
usually regards adult motives as variations upon one monotonous 
standard pattern. 

Hormic psychology , promising though it is in its recognition of 
sentiments, is cramped by preoccupation with the uniform instincts 
presumed to underlie these sentiments. 

Psychiatry and other practical arts of dealing with whole men come 
closer to adequacy, but as yet they are deficient in conceptual formu- 
lation, and do not advance the theoretical psychology of personality. 
Such formulations as they have are derived principally from the study 
of disease rather than health. 

Personalistic psychology also puts its emphasis on totality, but as 
yet its theories are broad and philosophical, failing to provide specific 
implements for bringing the concrete complexities of single personali- 
ties under inspection (cf. Chapter XX). 

V erstehende psychology is the only school of psychological 
thought that flatters human personality, finding it just as sublime as 
the ideal types produced in the minds of pre-Hitler German profes- 
sors. It thereby weakens its own effectiveness; and commits in addi- 
tion the fault of bifurcating psychology, refusing to utilize the results 
of any investigations other than its own (pp. 17, 231). 

This uncomplimentary survey does not deny that each of these 
branches of psychology has its own distinctive merit in the study of 
personality, but it does deny that any one of them is at present fully 
equipped to handle the problems of maturity in personality as defined 
in this chapter. The psychology of personality cannot at present come 
to roost under any one of them. In time some may expand and show 
greater adequacy (promising signs exist especially in the last five 
and in Gestalt theory). But the required progress will come only if 
criticisms concerning inadequacy zxt taken to heart. In the mean- 
time new trails must be blazed. A beginning may be made by return- 


ing now to further consideration of the three attributes of mature 
personality with which up to now psychology has so inadequately 


The sense of self built up so laboriously in infancy (cf. pp. 
158-165), sharpened and strengthened especially during the period of 
negativism, must not be regarded as fully formed in the first three, 
nor in the first ten years. It continues to expand with experience, 
with emotional involvements, frustrations, discriminative adjust- 
ments, and insight. Owing to the special psychic isolation typical of 
adolescence the sense of self at that time is exceedingly poignant. 
But neither does development stop here. The introversion of the self 
at this period of life merely prepares for further expansion later on. 

Take that most celebrated of all periods in life, the time of falling 
in love, when both the organization and the sense of self are extended. 
Falling in love condenses into one sharply personalized sentiment all 
sorts of previously unrelated dispositions: specific sexual tonicity, 
assertive and submissive tendencies, habits, ambitions, esthetic inter- 
ests, family sentiment, and often too religious interest and emotion. 
Most important is the fact that this intimate surge involves some 
other person. What is of interest to another becomes vital to oneself. 
Hitherto self-sufficient, the lover finds himself no longer so. The 
welfare of another is more important than his own. In this way the 
self is extended. 

And still the process goes on. Possessions, friends, one’s own chil- 
dren, other children, cultural interests, abstract ideas, politics, hobbies, 
recreation, and most conspicuously of all, one’s work, all lead to the 
incorporation of interests once remote from the self into selfhood 
proper. What one loves becomes a part of him. And anything one 
can admire, feel sympathy for, appreciate, revere, deliberately imi- 
tate, or become unconsciously identified with, may become intro- 
cepted into the personality, and remain ever after a vital part of it. 1 

1 The term introception > originated by W. Stern, stands for the adoption by an 
individual of cultural standards (conventions, morals, ideals) into his own personal 
system of motives and desires, or the incorporation of the interests and values of 
other human beings into his own life. The socialized person introcepts the standards 
of his group, the devout churchman introcepts the teachings of his faith. What was 
at first outer and perhaps alien becomes inner and dynamic. (Cf. AUgemeine Psy- 
chologies 1935, p. 102.) 

The concept of introception is not strictly speaking a psychological concept, 
but ethical, signifying the transformation of beierotelesis into miotehsh. The psy* 

2 l8 


Introspectively the self is extended and widened; objectively a per- 
sonality is evolved and matured. 

In the conversation of a truly mature personality his discourse 
does not seem to spring from biologically bound dispositions as much 
as from his acquired autonomous interests. It has been said cryp- 
tically that the sign of cultivation in a man is his ability to talk for 
half a day without betraying his occupation. Such demonstration of 
an extended range of interests is not necessarily the pose of superior 
breeding; if it be sham, as McDougall has pointed out, it is easy to 
detect; a solidly constructed system of sentiments, even to an un- 
trained observer, has a very different appearance from the mask of a 

Now Freud, as might be expected, gives a very different interpre- 
tation of the maturer interests of men: “The Super-ego answers in 
every way to what is expected of the higher nature of man.” Let us 
grant that a person is not mature unless he gives due respect to the 
codes of the society wherein he lives, acts with good taste and abides 
by the law, suffering pangs of conscience when he violates the rights 
of others and when remiss in his prescribed duties. But is this activity 
of the Super-ego all there is to the “higher nature” of a man? Not at 
all, for left to itself the Super-ego would produce a personality com- 
pletely caked with custom and shackled by tribal mores. Conven- 
tionality is not the same as maturity. 

The genuinely mature person has an Ego-ideal as well as a Super- 
ego. The former sets a goal that leads to a creative pattern of life, 
whereas the latter alone leads to static and stupid conventionality. 
It is a complete give-away of Freudian ideology that these two con- 
cepts are often considered synonymous. In psychoanalytic theory the 
poor Ego has no recourse but to surrender to one of its two tyrants, 
the Super-ego or the Id, or to compromise with both as best it may. 
The Ego-ideal, on the contrary, is the plan that the developed per- 
sonality is able to evolve for defeating, by transcending, both the un- 
socialized urges of the Id and the dullness of the Super-ego, leading 
thereby to a new level of personal freedom and to maturity. 

Intelligent and perspicacious planning for the future is always a 
significant feature of any mature life. The individual imagines things 
as they might be, even picturing his own personality as he would 

chological counterparts of introception are many, including imitation and all other 
forms of learning, traumatic effects, suggestion, identification, imagination, and above 
all (since introcepted values become motives) functional autonomy. 


like it. This planning for the future determines the subsequent de- 
velopment of personality quite as effectively as do the forces of the 
past. It is not only the vires a tergo that create a style of life, but 
also the plans, ambitions, ideals, and images that mediate goals pro- 
jected into the future. Every mature personality may be said to travel 
toward a port of destination, selected in advance, or to several re- 
lated ports in succession, the Ego-ideal always serving to hold the 
course in view. That which lies ahead in one’s life is at every moment 
dynamically taking shape, not merely by virtue of the push of this 
habit and that stimulus, but because the course of development is 
being steered in a certain direction by the Ego-ideal itself. 

The importance of this concept of directionality in mature per- 
sonalities is clearly brought out by the investigations of Dr. C. Biihler 
and her associates . 2 In their study of approximately two hundred 
life-histories, the most definite conclusion was that each life seemed 
definitely ordered and steered toward some selected goal; each person 
had something quite special to live for. Each had a characteristic 
Bestimmmg and intention. The style, of course, varied; some staked 
everything upon one single great objective; others varied their goals 
from time to time, but goals there always were. A supplementary 
study of would-be suicides showed that life becomes intolerable to 
those who can find nothing to aim at, no goal to seek. 

In childhood the objectives are at first lacking altogether; in 
adolescence they are vaguely defined; early maturity brings definite- 
ness of plan; and the remainder of the active years of life are spent 
in the pursuit. The Bestimmung is not, of course, unceasingly effec- 
tive. Sometimes it weakens, and often it is defeated by uncontrollable 
factors. Lives plagued by bad luck may be forced to alter their ob- 
jectives and to take a more modest goal (to lower their levels of 
aspiration) . Sometimes, on the other hand, there is grim persevering 
in the face of insuperable obstacles, a decision to continue in the 
selected road, quand mime. Some defeated personalities seem bound 
to life merely “by indignation,” but even this emotional focus serves 
as a goal for combat. 

It is perhaps a defect in Buhler’s study that she deals so exclusively 
with geniuses who of necessity are distinguished for their Bestim- 
vrnng. They could not have made their marks upon the world unless 

2 C. Biihler, Der menschlicbe Lebemlauf ah psychologists Problem, 1933; also 
E. Frenkel, “Studies in Biographical Psychology,” Char. & Pen., 1936, 5, 1-34. Ad- 
ditional details concerning these studies are given in Chapter XIV. 



they were endowed with unusual purposes, persistence, and vigor. 
But in the more ordinary lives of average people the principle may 
be found to operate in only slightly less degree. Unassuming lives 
revolve around unassuming foci of interest— a comfortable home, a 
routine vocation, perhaps the pursuit of health or of a health-cult. 
In such cases the degree of maturity may not be as marked as in the 
genius, but without some sustained goals somewhere, a personality 
remains childish. 


To live without self-deception has been the ideal of many. When 
Phaedrus, walking with Socrates, asked some question concerning a 
local legend, Socrates replied, “Now, I have no leisure for such en- 
quiries. Shall I tell you why? I must first know myself, as the Del- 
phian inscription says; to be curious about that which is not my 
concern, while I am still in ignorance of my own self, would be 
ridiculous.” 3 

Knowledge of oneself is called insight. The term is an extension 
of the psychiatric usage according to which a mental patient who 
knows that he (and not everybody else) is suffering from disorienta- 
tion and aberration, is credited with insight. Also in the domain of 
normal personality insight means freedom from self-deception. 4 It is 
the trait that Lord Chesterfield claimed for himself when he wrote to 
his son, “I know myself (no common piece of knowledge, let me 
tell you) . I know what I can, what I cannot, and consequently, 
what I ought to do.” 

When he wrote these self-confident lines Chesterfield probably 
did not realize that most human beings feel equally sure of their 
superiority in this same “uncommon” piece of knowledge. Others 
may lack insight, but not 1 ! In one study 96 per cent of the students 
in various courses in psychology thought that they possessed average 
or better than average insight, only 4 per cent admitting possible de- 
ficiency. Unlike Socrates they were not willing to profess ignorance 
of themselves. One feels that living with oneself naturally results in 
a thorough acquaintance. But such familiarity may also work in the 
opposite way, by hardening one to his defects of memory and intelli- 

3 Plato, Phaedrus, 229. 

4 This usage is not to be confused with two quite different psychological mean- 
ings of the term, viz., sudden learning (“learning by insight”), and clear compre- 
hension (“he has good insight into human nature”). 



gence, to his defenses and rationalizations, and to his own impulsive 
justifications of his deeds. 

Some people, to be sure, seem to make a point, even a virtue, of 
admitting their defects, perhaps writing their Confessions “ob- 
jectively” for the world to read. But the chances are that they guard 
some secret tabernacles against prying eyes, even their own. Perhaps 
it is nothing of major importance that is withheld, only an incident 
of meanness or shame; but it would be too humiliating to disclose, or 
even to face. 5 

The value of insight seems never to be questioned. No writer 
makes out any case for self-deception. Not infrequently insight is 
exalted to the highest place among the virtues, or therapeutically is 
regarded as a panacea for all mental ills. Now, admitting that insight 
is a prerequisite for almost any intelligent change in oneself, it still 
does not follow that self-knowledge is synonymous with virtue, as 
Socrates would have us believe. The transformation of personality 
does not take place automatically. What insight does is to make past 
mistakes intelligible so that one is not condemned through ignorance 
to repeat them. Furthermore insight removes needless worries by 
showing their groundlessness (unless through functional autonomy 
they have become firmly set). But for any basic change, insight must 
be supplemented by a new orientation, a vigorous plan for the future, 
a new and effective motivation. 

How is the psychologist to tell whether or not an individual has 
insight? According to an old adage, “Everyman has three characters: 

(1) that which he has, 

(2) that which he thinks he has, 

(3) that which others think he has.” 

Ideally, insight is to be measured by the ratio between the second 
item and the first, for what a man thinks he is in relation to what he 
really is provides a perfect definition and therefore an admirable 
index of his insight. Practically, however, proof positive of what a 
man is in the biophysical sense is difficult to obtain; ultimately, there- 
fore, the most practicable index of a man’s insight becomes the ratio 
between the second and the third items, the relation of what a man 
thinks he is to what others (especially the psychologist) think he is. 

6 An interesting discussion of the shortcomings of Confessions and other self* 
revelations as psychological documents is -given by Stefan Zweig, Adepts in Self 
Portraiture, 1928, po. xvi-xviii. 



If the man objects that all the world, including the psychologist, is 
'wrong about him, he cannot of course be disproved, but in such a 
case the evaluation of his insight becomes virtually impossible. 

Psychologists know that there are certain correlates of insight, 
qualities that people of good insight possess. For example, those who 
are aware of their own objectionable qualities are much less likely 
to attribute them to other people, that is, they are less given to pro- 
jection than are those who lack insight. 8 Also, people of good in- 
sight are known to be more Intelligent than the average. 7 

But the most striking correlate of insight is the sense of humor. 
In one study where a number of subjects rated one another on a 
large number of traits, the correlation between ratings on insight and 
humor turned out to be +.88, the highest in the whole series. Such 
a high coefficient means either that personalities with marked insight 
are also distinguished for their humor, or else that the raters were not 
able to distinguish between these two qualities. In either case the re- 
sult is important. Again one thinks of Socrates, and recalls the legen- 
dary association of these two qualities in his personality. It has been 
told how at a performance of Aristophanes’ Clouds he stood up in 
order that the amused audience might better compare his face with 
the mask that was intended to ridicule him. Having objectified him- 
self, he was able to view the caricature of his comic physique as an 
esthetic and detached event, and to aid the jest by laughing at him- 

The sense of humor must be distinguished sharply from the cruder 
sense of the comic. The latter is a common possession of almost all 
people, children as well as adults. What is ordinarily considered 
funny— on the stage, in comic strips, on the radio, or in ordinary 
life— consists usually of absurdities, horse play or puns. The laughter 
provoked in these cases has a very different explanation from amuse- 
ment due to subtleties of true humor. For the most part the comic 
consists in the degradation of some imagined opponent, as Aristotle, 
Hobbes, and many others have pointed out. Or it consists of the 
abrupt and sly release of some suppression, as in the case of risque 
stories. There may be other thematic elements (fear, aggression, 
hate), which cause laughter when people are permitted by the joke 
to vent some troublesome and semi-conscious tension. 8 Then there is 

8 Cf. R. R. Sears, /. Soc. Psychol, 1936, 7, 151-163. 

7 P. E. Vernon, /. Soc. Psychol, 1933, 4, 42-57. 

8 Cf. H. A. Wolff, C. E. Smith, and H. A. Murray, /. Abnorm , & Soc . Psychol , 
'934* 28, 341-365. 


the laughter of good spirits, easily provoked in children or in adults 
at play. But none of these forms of fun correspond to the humor 
we are speaking of. They are not related to insight. 

True humor has been defined by the novelist Meredith as the 
ability to laugh at the things one loves (including of course oneself 
and all that pertains to oneself), and still to love them. The real 
humorist perceives behind some solemn event, himself for instance, 
the contrast between pretension and performance. That which he 
values becomes, for the time being, vain show. There is a sudden 
shift of emphasis; for the moment all the world’s a stage where 
nothing really matters, and where the actors, including oneself, can be 
viewed with the detachment of Olympus. 

Humor of this type seems to have a development entirely parallel 
to that of insight. A young child lacks both. He cannot see himself 
as others see him, and his mirth is seldom directed toward himself. 
He laughs readily enough, but usually at the minor misfortunes of 
others. At this stage the degradation theory of laughter clearly ap- 
plies. In adolescence, insight is but rarely attained, not because the 
youth is as unmindful of himself as is the young child, but for the 
opposite reason, in his intense seriousness he lacks perspective. There 
are feelings of acute self-consciousness and of inferiority, largely be- 
cause a sense of proportion has not developed. His failures and ec- 
centricities do not amuse him; he is much more likely to weep about 
them than to laugh. In some adolescents, to be sure, this condition of 
storm and stress is much less marked than in others. 

Typically then, childhood is marked by freedom from self- 
examination, an implicit “I don’t care.” In adolescence, when per- 
sonal values first become deep and significant, there is a profound 
“I do care” attitude. But in maturity, a sensitive and intricate bal- 
ance is attained, peculiar to each life, between caring and not caring, 
between valuing and recognizing the vanity of value. Only at this 
mature stage can the individual both pursue his course diligently and 
at the same time apply the advice of Mark Twain’s Mysterious 

Your race in all its poverty has unquestionably one really effective 
weapon— laughter. Power, money, persuasion, supplication, persecu- 
tion— these can lift a colossal humbug-push it a little-weaken it a 
little century by century; but only laughter can Now it to rags and 



atoms at a blast. Against the assault of laughter nothing can stand 
You are always fussing and fighting with your other weapons. Do 
you ever use that one? No, you leave it lying rusting. 

To achieve a sense of humor as well as insight requires a high 
level of intelligence. It is only the most intelligent who prefer their 
humor objective and realistic, based on the exact relationships obtain- 
ing in their lives. People less intelligent prefer humor derived from 
their own repressions and reflecting marked thematic elements. 9 In- 
sight does away, in large part at least, with the dynamic power of 
these thematic elements. If one knows one’s inferiorities, jealousies 
and unsocial desires, one is less likely to take pleasure in their autistic 
triumph through a mere joke. It requires intelligence to see oneself 
in perspective, and like Socrates to be amused at one’s own pomposity 
and pretensions. 

The reason why the attainment of insight and of humor march 
hand in hand is probably because they are at bottom psychologically 
a single phenomenon— the phenomenon of self -objectification. The 
man who has the most complete sense of proportion concerning his 
own qualities is able to perceive their incongruities and absurdities 
in other than their customary, frames of reference. Yet perspective 
does not mean superciliousness. Values have their place, even though 
they be viewed at times in strange and incongruous settings where 
their pretensions and incongruities are disclosed. 

As with insight, almost everyone claims a sense of humor as his 
special and peculiar possession. The same students who were asked 
to evaluate their own insight in comparison with that of other people 
(cf. p. 220) were also asked to estimate their sense of humor. Ninety- 
four per cent replied that it was as good or better than the average! 
Stephen Leacock has observed the same conceit, reporting it in My 
Discovery of England: 

A peculiar interest always attaches to humor. There is no quality of 
the human mind about which its possessor is more sensitive than the 
sense of humor, A man will freely confess that he has no ear for 
music, or no taste for fiction, or even no interest in religion. But 1 
have yet to see the man who announces that he has no sense of humor 
In point of fact, every man is apt to think himself possessed of an ex- 
ceptional gift in this direction. ... 

9 Cf. C. Landis and W. H. Ross. L Soc. PsychoLj 1933, 4* 156*175. 


But, after all, it is perfectly natural for each of us to think him- 
self superior in both insight and in humor. One cannot readily admit 
in himself deficiencies that he cannot observe, nor can he readily con- 
cede that an event is beyond his sense of humor if he can see nothing 
funny in it. If he could, he would already have insight and, no doubt, 
be amused by the whole situation! 


Humor in one respect, but in one respect only, is like religion. 
By setting a frame of reference that is at variance with the ordinary 
mundane frame of reference, both have the peculiar ability of pre- 
cipitating the ordinary worries and mischances of life into new and 
sane patterns. Humor, like religion, shatters the rigidity of literal- 
; mindedness. To view one’s problems humorously is to see them as 

trivial and of no consequence; to view them religiously is to see them 
\ ' in relation to a divine scheme that gives them changed meaning. In 

t humor things are not at all earnest or purposive, but pompous and 

out of step; in religion there is no such thing as incongruity. Thus 
i setting up novel standards, both religion and humor, albeit in very 

different ways, bring perspective. 

Since the events of life cannot possibly be regarded at any one 
j time as of great moment and as trivial, it follows that a personality 

; cannot be at one time both reverent and jesting. Yet at bottom 

■ the mature person may be profoundly religious and still have the 

| capacity for humor. He may even joke and pray about the same 

j disturbing events in his own life— though never at the same time, 

i At bottom he is an absolutist, but a certain delicate balance of inter- 

J ests makes it possible for him also to be on occasion as fun-loving 

I as any pluralist. Max Eastman depicts the situation: 

I Mahomet boasted that with faith and prayer he could make a moun- 

tain get up and come to him. And when a great crowd of his follow- 
ers had assembled, and all his incantations failed, he said: “Well, if the 
f mountain will not come to Mahomet, Mahomet will go to the moun- 

) tain.” And so we strive with all our energy and ingenuity to make the 

course of things do us pleasure, and as the course of things continually 
disappoints us, we say: “Very well, I will find a special pleasure in the 
;; disappointment!” That is our sense of humor. Humor is Mahomet 

going to the mountain . 10 

f 10 T he Sense of Humor, Charles Scribner’s Sons, lgzz, p. zj f. 



And yet what keeps the religious person from becoming a cynic- 
as all thoroughgoing humorists must be— is the conviction that at 
bottom something is more important than laughter, namely, the fact 
that he, the laugher, as well as the laughter itself, has a place in the 
scheme of things according to the dispensations of a Divine Intelli- 
gence. When this most important of issues is decided, there is still 
plenty of room for jesting. In fact a case might be made for the poten- 
tially superior humor of the religious person who has settled once and 
for all what things are of ultimate value, sacred and untouchable, for 
then nothing else in the world need be taken seriously. He can readily 
concede that the bulk of worldly happenings are ludicrous, that men 
and women, including himself, are given to amusing vanities, actors 
on a stage set with human artifices. To him nothing in their coming 
and going is of consequence, except their ultimate salvation. Most 
things they do are merely laughable. Beyond the reach of humor lies 
one and only one serious purpose to which humor must give way 
whenever the two are in conflict . 11 

Religion is the search for a value underlying all things, and as 
such is the most comprehensive of all the possible philosophies of life. 
A deeply moving religious experience is not readily forgotten, but is 
likely to remain as a focus of thought and desire. Many lives have no 
such focus; for them religion is an indifferent matter, or else a purely 
formal and compartmental interest. But the authentically religious 
personality unites the tangible present with some comprehensive view 
of the world that makes this tangible present intelligible and accept- 
able to him. Psychotherapy recognizes this integrative function of 
religion in personality, soundness of mind being aided by the posses- 
sion of a completely embracing theory of life. 

Besides the religious, there are many other unifying philosophies. 
Though less embracing in their scope, they too serve as autonomous 
systems wherein every detail tends to corroborate every other detail 
under some fundamental conception of value. Take for instance the 
esthetic philosophy of life. The poet says of himself, 

For beauty have I striven, 

My blood flowed in my song. 

11 It is only the core and aim of a religious faith that are beyond the 
reach of humor. All human foibles related to the religious intention are possible 
sources of amusement, examples being the incongruities due to human failing in the 
act of worship. Plenty of amusing episodes occur in church. It is only the ultimate 
aim of the acts of worship that, for the religious person, lies beyond the scope of 


All else for him is ordered under this one value and derives im- 
portance from it. He does not mean that he always achieves beauty, 
that he does not at times produce ugliness, nor that he cannot laugh 
at himself and his failures. He does mean that neither incompetence, 
backsliding, nor mirth, are able to defeat permanently this one unify- 
ing idea. It is the dynamic idea, not his skill with its mortal failings, 
that unifies his life. 

Such a view contrasts sharply with the genetic over-emphasis of 
other dynamic psychologies that regard motivation as sessile to the 
roots of the past. Take striving for poetic beauty. One psychoanalyst 
claims that far from being a self-sufficient interest, love of poetry is 
simply an expression of oral eroticism, “a chewing and sucking of 
beautiful words and lines.” 12 Hence this poem is not a forthright and 
honest account of its author’s philosophy. His true philosophy is 
hidden from him; it consists of nothing more than the desire to secure 
oral gratification by sucking. The same dispute is in process regard- 
ing the nature of religious interests. Freud declares himself “perfectly 
certain” that this particular class of “illusions” springs from infantil- 
ism of the mind. 13 

Which interpretation is correct? Are esthetic and religious phi- 
losophies of life due to a flatulent condition of the Id that “never 
changes”; or are they precisely the opposite, autonomous master- 
sentiments that give objective coherence and subjective meaning to 
all the activities of their possessors’ lives? By now the reader is in a 
position to decide for himself. 

By and large psychology has done little to give systematic setting 
to all these various dynamic formations that represent the apex of 
development in the mature personality. With time, no doubt, when 
the errors of excessive elementarism and geneticism are cleared away, 
and the principle of functional autonomy is substituted as a general 
guide, the situation will improve. 

In the meantime the one school of psychological thought that has 
concerned itself exclusively with these phenomena, and actually suc- 
ceeded in flattering human nature (a thing psychology seldom does) 
is the school of V erstehendepsychologie of Dilthey, Spranger, Jaspers, 
and other German authors. 

The method of this school is strikingly original. It consists in the 
postulation of ideal types, representing ultimate and absolutely co- 

12 A. A. Brill, Imago, 1933, 19, 143-167. 

13 S. Freud, New Introductory Lectures on Psychoanalysis , 1933, p. 225. 



herent patterns of value, unifying any personality capable of follow- 
ing one of them consistently. Though V erstehendepsy chologie ad- 
mits that no individual is perfectly self-consistent (the types repre- 
senting merely “schemata of comprehensibility” to aid in under- 
standing), still the effect of the method, since it portrays only ideal 
types, is one of flattery. 

The various authors of this school each present different ideal 
types, the best known being the six of Spranger . 14 Two of these value- 
directions, the religious and the esthetic , have already been mentioned 
as common instances of unifying philosophies of life. The four others 
completing the list are the theoretical, the economic, the social, and 
the political. The list is purely a priori, but it provides one of the few 
available hypotheses for the empirical psychologist interested in the 
question of master-sentiments . 16 The following summary describes 
the six ideal types as Spranger sees them . 16 

1. The Theoretical The dominant interest of the theoretical man is 
the discovery of truth . In the pursuit of this goal he characteristically 
takes a “cognitive” attitude, one that looks for identities and differ- 
ences; one that divests itself of judgments regarding the beauty or 
utility of objects, and seeks only to observe and to reason. Since the 
interests of the theoretical man are empirical, critical, and rational, he 

i . is necessarily an intellectualist, frequently a scientist or philosopher . 17 

Flis chief aim in life is to order and to systematize his knowledge. 

2. The Economic . The economic man is characteristically interested 
in what is useful Based originally upon the satisfaction of bodily needs 
(self-preservation), the interest in utilities develops to embrace the 
practical affairs of the business world— the production, marketing and 
consumption of goods, the elaboration of credit, and the accumulation 
of tangible wealth. This type is thoroughly “practical” and conforms 
well to the prevailing conception of the average American business 

The economic attitude frequently comes into conflict with other 
values. The economic man wants education to be practical, and re- 
gards unapplied knowledge as waste. Great feats of engineering, sci- 

14 E. Spranger, Types of Men, trans. 1928. 

15 The empirical validity of these types is discussed in Chapter XV. 

16 E. Spranger, op. cit., , pp. 109-246, 37-106, 319-347. The condensed description 

P iven here is reprinted from P. E. Vernon and G. W. Allport, /. Abnorm . & Soc . 
sychoL , 1931, 26, 231-248. 

17 It must not be thought that a high degree of talent or attainment is necessary 
to qualify a person for classification in this, or in any, type. Even the most undis- 
tinguished personalities are to be known not by their achievements but by their in- 
terests and intentions. 


entific management, and “applied psychology” result from the de- 
mands which economic men make upon learning. The value of utility 
likewise conflicts with the esthetic value, excepting when art serves 
commercial ends. Without feeling inappropriateness in his act, the 
economic man may denude a beautiful hillside or befoul a river with 
industrial refuse. In his personal life he is likely to confuse luxury 
with beauty. In his relations with people he is more likely to be in- 
terested in surpassing them in wealth than in dominating them (politi- 
cal value) or in serving them (social value). In some cases the eco- 
nomic man may be said to make his religion the worship of Mammon. 
In other instances, however, he may have regard for the traditional 
God, but inclines to consider Him as the giver of good gifts, of 
wealth, prosperity, and other tangible blessings. 

3. The Esthetic . The esthetic man sees his highest value in form and 
harmony . Each single experience is judged from the standpoint of 
grace, symmetry, or fitness. He regards life as a manifold of events; 
each single impression is enjoyed for its own sake. Fie need not be a 
creative artist; nor need he be effete; he is esthetic if he but finds his 
chief interest in the artistic episodes of life. 

The esthetic value is in a sense diametrically opposed to the theo- 
retical; the former is concerned with the diversity, and the latter with 
the identities of experience. The esthetic man chooses with Keats to 
consider truth as equivalent to beauty, or else to agree with Mencken, 
that “to make a thing charming is a million times more important than 
to make it true” In the economic sphere the esthete sees in the process 
of manufacturing, advertising, and trade, a wholesale destruction of 
the values most important to him. In social affairs he may be said to 
be interested in persons but not in the welfare of persons; he tends 
toward individualism and self-sufficiency. Esthetic people often like 
the beautiful insignia of pomp and power, but oppose political activity 
when it makes for a repression of individuality. In the field of religion 
they are likely to confuse beauty with purer religious experience. 

4. The Social. The highest value for this type is love of people, 

whether of one or many, whether conjugal, filial, friendly, or phil- 
anthropic. The social man prizes other persons as ends, and is there- 
fore himself kind, sympathetic, and unselfish. He is likely to find the 
theoretical, economic, and esthetic attitudes cold and inhuman. In 
contrast to the political type, the social man regards love as itself the 
only suitable form of power, or else repudiates the entire conception 
of power as endangering the integrity of personality. In its purest 
form the social interest is selfless and tends to approach very closely 
to the religious attitude. L Y ■ 

5. The Political The political man is interested primarily in power. 


His activities are not necessarily within the narrow field of politics; 
but whatever his vocation, he betrays himself as a Machtmensch. Lead- 
ers in any field generally have high power value. Since competition 
and struggle play a large part in all life, many philosophers have seen 
power as the most universal and most fundamental of motives (cf. p. 
169). There are, however, certain personalities in whom the desire for 
a direct expression of this motive is uppermost, who wish above all 
else for personal power, influence, and renown. 18 

6. The Religious. The highest value for the religious man may be 
called unity. He is mystical, and seeks to comprehend the cosmos as a 
whole, to relate himself to its embracing totality. Spranger defines the 
religious man as one “whose mental structure is permanently directed 
to the creation of the highest and absolutely satisfying value experi- 
ence.” Some men of this type are “immanent mystics,” that is, they 
find in the affirmation of life and in active participation therein their 
religious experience. A Faust with his zest and enthusiasm sees some- 
thing divine in every event. The “transcendental mystic,” on the 
other hand, seeks to unite himself with a higher reality by withdraw- 
ing from life; he is the ascetic, and like the holy men of India, finds 
the experience of unity through self-denial and meditation. 

One advantage of such portraits (even though they are too per- 
fect in consistency to exist in real life) is the antidote they provide 
against excessive emphasis upon genetic factors and segmental 
analysis. Spranger’s classification offers a starting point for empirical 
investigations of those complex philosophies of life that serve more 
than anything else to confer unity upon the mature personality. 

There are several reasons why the guidance of this particular 
school of thought cannot be followed in every respect. In the first 
place, its preoccupation with the ideal exalts human nature too highly; 
the types are obviously exaggerations. For a better balance the whole 
doctrine must be translated into empirical terms. Since the various 
authors do not agree on the number or the nature of these types, 
they can be considered at best to represent only some characteristic 
philosophies of life. There are others that might be named, especially 
of the more earthy type. Hedonistic, sensual, and vital values are the 

18 The following telegram sent by Mussolini to the American ambassador on 
the occasion of Lindbergh’s flight to France, shows the manner in which the rou- 
tine acts of a person are influenced by his dominant values. Note the number of 
expressions indicative of Mussolini’s interest in power: “Please accept the shouts of 
enthusiastic admiration which at this moment ring from the hearts of all the people 
of Italy exulting over the superb oceanic flight by Lindbergh. A superb human 
wall took by assault space and subjugated it. Matter once more bowed to the mind 
and the wizard [ry] of men, to the glory of Lindbergh and his people.” 


2 3i 

foci of development in many personalities, but neglected by Spranger. 
Furthermore, these Lebmsformen are at best only categories of value 
There are many individual patterns possible within each, many kinds 
of social interests, for example, each producing a different course ol 
development in personality. All typologies are too broad, and this is 
no exception. Individuality becomes lost in such coarse classification. 

Finally, this school of the Geistesnvissenshaften is irreconcilably 
antagonistic to the contributions of all other branches of psychology, 
refusing not only to admit their divergent theories, but also their cor- 
rective empiricism. Such aloofness is fatal. Though each branch ol 
psychology considered by itself has limitations, their combined re« 
sources must be pieced together critically to give even an approxi- 
mately adequate account of the development of personality. 

Part III 




T he progress of any science, it is said, depends in large part upon 
its ability to identify elements which, in the combination found in 
nature, constitute the phenomenon that the science has set out to 
examine. Without its table of elements, chemistry could not exist; 
without the discovery of the cell, biology would be little better off 
than primitive animism; physics, while it has discarded one set of ele- 
ments after another, comes again in the quantum theory to another 
type of element. 

The suspicion with which many natural scientists view psy- 
chology arises in part from their belief (entirely correct) that the 
elementary processes of mind have not yet been identified. Not that 
psychology has neglected the search for basic units with which to 
work. It has embraced hopefully many possibilities: faculties, ideas, 
instincts, reflexes, sensations, images, affects, factors, dimensions, and 
others— but the shifting of the lists, according to the predilections of 
the various investigators, has prevented a common meeting ground. 
Partly as a result of this failure to agree there has set in a reaction 
against the search for “mental atoms.” The doctrine of the whole 
has won popularity. Yet the search for units has not in fact been 
abandoned. Only a shift of conception regarding the nature of the 
elements has occurred. Psychic atoms are denied, but psychic struc- 
tures are affirmed, and these structures turn out to obey certain 
principles of organization, and to be composed in turn of sub- 
structures, discoverable through the process of orderly analysis. And 
so even in its modern configurational phase, psychology is engaged 
as ever in the same difficult search for unit structures. For without 
some guiding hypothesis concerning the most suitable level of analy- 
sis, only wavering and ambiguous results can be achieved. 

The present chapter, first in a series devoted to this pivotal scien- 
tific problem, surveys the principal kinds of elements thus far pro- 
posed by psychologists as methods for “breaking down” personality 
into units suitable for purposes of comparison, measurement, and de- 




A psychologist or psychiatrist faced with the practical task of 
counseling or therapy sometimes devises a schedule of components 
of personality for his own guidance in dealing with his clients or 
patients. The result is a list of distinguishable aspects of personality 
that he considers fundamental Since the counselor uses the same list 
for all personalities, he is committing himself to a uniform schedule 
of elements; to all cases he applies the same rubrics. But such classi- 
fications are usually nondescript, consisting in a potpourri of units 
without a thought as to what the essential nature of a unit may be. 
Convenience and practical utility are the only guides; there are no 
other principles underlying the selection. A few representative cita- 
tions will illustrate this kind of classification. 

The Eugenics Record Office has issued a Trait-Book containing a 
list of approximately 3,000 characteristics that might conceivably be 
hereditary according to the principle of unitary characteristics. 1 But 
the non-comparable nature of the items shows the lack of underlying 
psychological or genetic theory. What logic of elements could pos- 
sibly apply to such a mixture as albinism. , eczema, love of fishing , ag- 
gressiveness, religiosity, and ingrowing toenails? 

Of a different order, but still nondescript in composition, are the 
numerous guides to self-study, or charts for aiding in the complete 
description of a personality. 2 One and all these schemes have a prac- 
tical aim; they disclose aspects of personality that a more casual ap- 
proach is not likely to discover. But they are so diverse, they overlap 
in so few points, and are so seldom supported by any theory of the 
relation of the chosen units to one another, that they contribute noth- 
ing to the search for a suitable logic of elements. 

1 Eugenics Records Office (N. Y.: Cold Spring Harbor), The Trait-Book , 1919, 

2 There are many such available guides, decidedly useful in the descriptive study 
of personality, but not consistent even with themselves in their conceptions of the 
nature of the structure of personality. G. S. Amsden, A Guide to the Descriptive 
Study of the Personality, 1924 (N. Y.: Bloomingdale Hospital Press); F. H. Allport, 
Systematic Questionnaire for the Study of Personality , 1925 (Chicago: Stoelting 
Co.); J. O. Chassell, The Experience Variables Record , 192B (Rochester: Univ. of 
Rochester Medical School); F. L. Wells, “The Systematic Observation of Per- 
sonality,” Psychol Rev,, 1914, 21, 295-332; W. Baade, O. Lipmann, W. Stem, “Frag- 
ment eines psychographischen Schemas,” Zsch, f. ang, Psychol, 1909, 3, 191-315; 
G. Heymans and E. Wiersma, “Beitrage zur speziellen Psychologic auf Grund einer 
Massenuntersuchung,” Zsch. f , Psychol, 1906, 42, 81-127, 253-301; 1919, 80, 76-89; A* 
Huth, Exakte Personlichkeitsf or s chung, 1930; A. Lasurski, Vber das Stadium der 
Individualitdt, 1912. 


At a higher level of complexity are certain schemes setting forth 
alleged major divisions of personality. Thus for McDougall, personal- 
ity is constituted of five major components: disposition , temperament , 
temper, character, and intellect . 3 Since any one of these factors is 
assumed to vary independently of the others, a doctrine of five initial 
complex units results. A conception similar in principle but wholly 
different in the selection of components, is that of Klages who holds 
the major constituents of personality to be sensation , motion, appre- 
hension, will, contemplation, and expression . 4 Each may be studied by 
itself, for each is fundamental A similar level of analysis is offered by 
l Beck in his conceptualization of the four factors in personality that 
^jay be studied with the aid of the Rorschach test: form perception , 
organizing energy , affective drive, and creative activity . 5 Taken to- 
gether these factors are thought to constitute the whole of personal- 
ity. The curious feature of these coarse analyses is that in each case 
the rubrics chosen are claimed to cover the entire personality. Yet 
there is between the schemes scarcely any overlap! 

Boven makes a similar attempt. He distinguishes three broad divi- 
sions or assises of personality . 6 These are: dispositions (innate tem- 
peramental factors), traits (acquired emotional attitudes), and linea- 
ments (acquired philosophical attitudes). The dichotomies implied in 
this sequence between what is native and what is acquired, between 
what is emotional and what is intellectual, make the scheme artificial 
from the start. • 

Such practical and a priori conceptions as these show that much 
of the search for elements in personality ir ^ product of solitary 
thinking. Each writer dives into the recesses of his own mind and 
pulls up what for him is a useful .scheme of analysis, but one unlike 
that of any other writer. The units chosen by any one author are 
nondescript enough, but when they are compared with the units 
chosen by other authors, the result is complete darkness and con- 


The justification of any scheme of analysis is always to be found 
in the purpose for which the analysis is made. A system of elements is 
“true” in so far as it fulfills the avowed intention of the analyst. The 
principal reason why psychologists do not agree with one another 

s W. McDougall Energies of Men , 1933, chap. xxiv. 

4 L. Klages, The Science of Character , trans. 1928, 

5 S. J. Beck, Amer. J, OrthopsychiaL , 1933, 3, 361-375. 

6 W> Boven, La Science da Car act ere, 1931. 



in their lists of elements is that each is animated by a slightly different 
intention. Until the purpose of an analysis and the psychologist’s aim 
are clearly specified (as they seldom are) it is not possible to argue 
about the suitability of one set of elements or another. For certain 
purposes it is fitting to view mind as a congeries of ideas, for other 
purposes, as a network of neural arcs, or as a system of vectors, or as 
an hierarchy of sentiments. Mind is capable of being all things to all 
psychologists according to their personal lines of interest. 

Yet where personality (mind-in-particular) is concerned the 
criteria involved in the selection of elements become narrowed. This 
is especially true if investigators subscribe to a limiting definition of 
personality. Suppose, for example, that they agree with this volume 
in considering personality as always distinctive and unique, as pos 
sessing strongly individualized motivational systems, and as repre- 
senting an individual’s style of adjustment and mastery within his be- 
havioral environment. If. the intention of the investigator is to depict 
with maximum fidelity the structure of personality as here defined, 
then certain classes of elements will be far more suitable than others. 
Let us consider some of the possibilities. 

Faculties. The oldest theories of the composition of personality 
were all in terms of faculties. Well into the nineteenth century, the 
differences between men were ascribed to the varying strength of 
the “Powers of the Mind” (Attention, Will, Sagacity, and the like). 
The mere naming of these powers seemed to the early psycholo- 
gists to render them somehow fixed and self-active. 

It will be noted that w doctrine of faculties assumes the ele- 
ments of personality to be the same as the elements of mind-in- 
general. The belief is that by combination these universal faculties 
will produce in action all possible expressions of personality. Even 
Gall who deliberately sought characterial faculties to replace the 
more intellectualistic faculties of his predecessors, kept this belief. 
Faculties for him were uniform for all men. They were also innate 
and independent of one another. So many indefensible assumptions 
are here involved that neither Gall’s guidance nor that of any other 
faculty psychologist can be accepted (cf. pp. 81-85). 

The Elements of General Psychology. When in the eighteenth 
century the French Sensationalists conceived the person to be the 
sum-total of his sensations, they were postulating the units of person- 
ality in terms of general psychology. Most psychologists today do 
precisely the same. For some, personality is a bundle of conditioned 


reflexes, for others the sum-total of a man’s habits, or his pattern of 
images and feelings, or the configuration of some other dimensions 
or factors converging in one organism. According to his own habits 
of thought, each psychologist tends to think of individuals as com- 
binations of whatever abstractions he happens to favor for psycho- 
logical analysis. 

This procedure, common as it is, is wholly unsuitable for the 
psychology of personality. For one thing such abstract units are not 
distinctively personal. Nor do they give any clue as to what kind of 
organization unites them into a concrete pattern of a single life. 

Dynamic Elements of the Nomothetic Order. When the irrational 
force of “instinct” and impulse in mental activity became fully recog- 
nized, the elements of personality were sought in urges or desires. 
But though the complexion of the elements was thus altered, the 
principle remained the same. Personality was now considered to be 
the individual pattern of instincts and wishes, forces of the Id and 
Super-ego, or of desires and needs. All these doctrines hold in com- 
mon the belief that men are formed in the same mold. There must 
be, so it is assumed, a finite number of stable forms of motivation, 
just as there is a finite number of chemical elements. These theories 
deny the possibility of an infinite number of goals toward which 
people strive, and an unlimited number of ways in which these goals 
are achieved . 7 

If it were not possible to approach still closer to the unique sub- 
structures of the individual personality, this class of theories would 
lead to the most progress. The nomothetic doctrine of needs as de- 
veloped by Murray is especially promising . 8 This author maintains, 
entirely correctly, that “No therapeutist, or indeed, anyone who has 
to deal in a practical way with human beings, can get along with- 
out some notion of motivational force (instinct, need, drive, impulse, 
urge, inclination, wish, desire, or what not).” Of these alternatives 
he selects for emphasis the steed which he regards as a directional 
force within the organism. When a need becomes active, a character- 
istic trend of behavior will usually ensue, even in the absence of the 
customary stimuli. Conversely, unless a need is present, responses to 
specific stimuli do not occur. Needs are recurrent, and whenever 
active produce a stress towards equilibrium. If suppressed they initiate 
the phenomena of displacement and fantasy described in Chapter VI. 

*Cf. G. Murphy, Amor. ]. Orthopsychiat., 1932, 2, 315-334. 

*H. A. Murray. /. Psychol., 1936, 3, 27-42. 



In some respects this doctrine is like the more familiar psychology 
of instincts, though it has the great advantage of being free from 
nativistic commitments. 

Since behavior is an important part of personality the science of 
personology cannot advance much further without a classification of 
the more important trends of behavior, or needs. In constructing such 
a classification, however, it is not necessary to limit oneself to needs 
which appear to be inherited. Which of the needs are innate, and to 
what extent, is another question— one for further observation and 
experiment . 9 

Murray enumerates several basic needs, in respect to which he 
believes all people may be profitably compared. The following are 
illustrative of his list: 

















Such a provisional classification has considerable value. It is heuris- 
tic; it yields orderly hypotheses that can be examined one at a time. 
It centers research, and gives coherence to the interpretation of data. 
Though free from the rigidity of instinct-doctrines, the approach is 
dynamic and places suitable emphasis upon the goal-directedness of 
behavior. It leads to flexible and penetrating depictions of person- 
ality . 10 

However, certain criticisms of this analytical scheme are in order, 
in so far as it is intended to serve as a theory of the structure of per- 
sonality. Needs are uniform elements and few in number; to them 
must be traced all motives of all men. The theory says, in effect, that 
objects of desire may vary from person to person, but the kinds of 
desire do not. Men may want different things, but there are only a 
few reasons why they want them. Two men, for instance, may be 
animated by a strong need for abasement; one perhaps becomes a 
sexual masochist, the other a well-disciplined monk. Does it not seem 

8 H. A. Murray, J, PsychoL, 1936, 3, p. 42. 

*° See the forthcoming volume. H. A. Murray, et aL } Explorations in Personality . 


unnecessarily abstract to assume one common need in these contrast- 
ing cases? Both men, to be sure, wish to abase themselves, but still 
a world of difference lies between their respective dispositions. Even 
the admission that other needs may be simultaneously present to alter 
the desires of each man, does not yield a concrete and life-like picture 
of his motives. Universalized needs fail to depict with exactness the 
special foci of organization existing in each individual life. Desire is 
always integral with its object, and its resulting forms are far more 
varied than such a limited list of needs would allow. In short, needs 
are disembodied and de-personalized to a greater degree than is justi- 
fied in elements that are to serve as the radicals of personality. 

Not only are needs conceived as separable from the objects 
sought, but as separable also from the capacities invoked in attaining 
these objects, and separable even from the shadings of taste and 
predilection that in real life are integral with all motivation. Here 
again they are too abstract. One cannot, without considerable vio- 
lence, distinguish the direction of the motive from the equipment it 
uses in skill and habits, from its attendant attitudes and tastes, and 
from its object-attachment. According to Murray, the need employs 
varying “actones” (modi opercmdi of adjustment, peculiar to the indi- 
vidual or to the specific occasion) . But is such a separation between 
end and means possible? Does any person have an abstract need for 
abasement? Is he not likely rather to be driven by an integral desire 
to serve such and such a cause, to obtain a special form of masochistic 
gratification, or to surrender himself to a certain type of religious 
experience? The need is not one for abasement in the abstract, but 
for a special, personalized form of self-surrender, in which the 
actones play an integral part with the need. (Sometimes, to be sure, 
the felt need is vague in respect to its goal-object; one feels that one 
would like to surrender oneself to something but does not know' pre- 
cisely how and wherefore. But even such vagueness of desire is still 
personalized. The attendant actones, images, goals, such as they are, 
though not clearly defined, are integral with the need. Vagueness of 
motive is not, after all, the same as abstractness of motive.) 

The point seems to be admitted when Murray writes, “Every 
need is associated with traces (or images) representing movements, 
pathways, agencies, goal objects, which, taken together, constitute a 
need integrate .” 11 Now, this conception of a need integrate is a great 

11 H. A. Murray, /. Psychol., 1936, 3, 37. 



improvement over the skeleton need. It fulfills well our demand for 
a unit of analysis that is concrete, life-like, and personal, provided 
only that the need integrate is understood to be not merely a momen- 
tary organization but a mental structure that endures and is a con- 
stant characteristic of the person. It is no doubt true that comparisons 
between individuals cannot be made quite so readily on the basis of 
individual need integrates as on the basis of common needs; but com- 
parison is only a secondary goal of the psychology of personality. 
The primary goal is the representation of the single life with maxi- 
mum fidelity. 

Factors. For the past decade a conception of elements widely 
favored by psychometricians has been that of factors. Applied first 
by Spearman to the search for components of intelligence, the fac- 
torial approach has been varied and improved, and is now frequently 
applied to the problem of the composition of personality. No one 
seems seriously to have questioned whether intelligence, an obviously 
artificial construct of the nomothetic order, might, in principle, offer 
a different type of scientific problem from the vital and integrated 
personality. No doubts of this order are raised; it is simply assumed 
that if general and special factors can be identified as the constituents 
of intelligence (that is, if certain common factors are found to ac- 
count for the inter-correlation of various tests purporting to measure 
intelligence), then an attempt should be made to reduce personality 
also to a few basic and uniform components, shared by all men. The 
search has already been carried out in many directions. There are 
attempts to isolate the basic patterns of vocational interests, for exam- 
ple, and to discover a few basic descriptive adjectives that can be 
applied to personality as designations of the fundamental ways in 
which men may differ, displacing the thousands of descriptive terms 
now loosely applied. Commonest of all are the applications of factor 
analysis to assorted batteries of “personality tests” in order to see what 
“unitary” factors will emerge from them. 

The technical formulae employed in factor analysis and their 
modes of application are far too specialized for discussion here . 12 It 
is necessary, however, to consider the kind of element that factor 
analysis offers as the root component of personality. 

A factor is an empirical, a posteriori construct or generalization. 

12 Instructive explanations of the aims and techniques of factor analysis may 
be found in L. L. Thurstone, The Vectors of Mind , 1955, and J, P, Guilford, Fsy- 
chometric Methods, 1936, chap, xiv. 


It is defined completely by its mode of derivation. Starting always 
with a battery of tests or ratings, heuristically presumed to cover the 
function that is the object of study, the factorial description of these 
tests reduces the scores of all the diverse items to the smallest number 
of independent variables. The overlap is consolidated, and the number 
of non-correlating factors discovered is assumed to provide the list 
of elements basic to the function. 

To take an illustration, suppose the topic of irritation or annoy- 
ance is the one the investigator wishes to study. He administers to a 
large population of subjects a questionnaire containing a wide range 
of pertinent items, hundreds of them, presumed in some way to be 
diagnostic of “annoyability.” One individual may report, for example, 
that he can’t endure people who sniffle, and that he is annoyed also 
by falling dandruff, smutty stories, crooked teeth, and aggressive 
salesmen. Other individuals likewise indicate their personal aversions. 
With suitable statistical treatment it turns out that when all the 
annoyances of all the people are correlated with one another, they 
are found to fall into certain major clusters. These clusters are opera- 
tionally defined by the component items themselves. Since it is normal 
and convenient to name these clusters, they are finally christened by 
the investigator with whatever generic terms seem to him best suited 
to embrace the component items. In the case of annoyances, one in- 
vestigator finds that the basic factors seem to be irritation at irregu- 
larities in personal appearance, at violations of the mores, offenses 
against one’s own ego, and annoyance at minor mishaps. 13 

The merits of such a procedure, extended to all the areas of per- 
sonality, are said to be great. For one thing, it is believed that a finite 
number of variables will be discovered, so that personality can at last 
be reduced to a schedule that might resemble the periodic table in 
chemistry. If basic elements of this order are discovered, then and 
only then can scales be invented for the accurate measurement of 
personality. Instead of relying on homemade devices developed by 
solipsistic ingenuity, scales would be developed for common use by 
all psychologists, measuring the same components in all personalities. 
At last a common psychological yard-stick would be available. A 
man’s psychograph, constructed of scores on these “essential traits of 
mental life,” would be directly comparable with the psychograph of 
every other personality. Finally, it is said, rather speculatively to be 

13 C. M. Harsh, An Inventory Study of Categories of Annoyance (Unix. Cali£ 
Library) , *935. 


2 44 

sure, that perhaps the factors eventually discovered— since they are 
common to all men— will be found to correspond in some way to the 
genetic units governing inheritance, and to basic and uniform struc- 
tures within the human nervous system. 

But there are difficulties with this conception of elements, enough 
difficulties to disqualify it under the theory of personality advanced 
in this volume. 

1. The initial assumption of factor analysis, shared by all other 
nomothetic theories, is open to challenge. Is it reasonable to assume 
that all people (or even those belonging to one “type”) do in fact 
possess the same basic constitution of personality? Must the foci of 
organization in all lives be the same? Must the factors, excepting for 
their differential weighting, be identical? 

An entire population (the larger the better) is put into the 
grinder, and the mixing is so expert that what comes through is a 
link of factors in which every individual has lost his identity. His dis- 
positions are mixed with everyone else’s dispositions. The factors thus 
obtained represent only average tendencies. Whether a factor is really 
an organic disposition in any one individual life is not demonstrated. 
All one can say for certain is that a factor is an empirically derived 
component of the average personality, and that the average personality 
is a complete abstraction. This objection gains point when one reflects 
that seldom do the factors derived in this way resemble the disposi- 
tions and traits identified by clinical methods when the individual is 
studied intensively. 

2. A second difficulty arises in the naming of factors. When the 
non-correlating clusters are isolated, they are often found to contain 
a most curious mixture of items, making no psychological sense. One 
such cluster, for example, contains the following hodge-podge: special 
acuities and pulchritude, combined with drive, but having some nega- 
tive relation to empathy and to spacial facility! Can any conceivable 
psychological sense be made of it? It is not always true that such 
meaningless mixtures are obtained, but the composition of any factor 
is likely to give some trouble in the selection of a suitable name. It is 
no solution to this problem for the investigator to resort to the com- 
promise of an initial letter, •w, p, o, s, e, or m, as though he did not 
quite dare pronounce the names of his factors out loud. It seems easier 
and more fashionable to defend an abstract symbol than to argue 
boldly for such substantial elements as •will, perseveration , oscillation, 


social ‘withdrawing , emotionality, and masculinity. And yet, if this 
modern version of faculty psychology has virtue at all, it must lie in 
its ability to identify the true and fundamental components of per- 
sonality which it set out to find. Up to now it has proved difficult to 
translate the confident products of mathematical derivation into an 
equally self-assured language of theoretical psychology. In brief, fac- 
tors often seem remote from psychological fact, and as such they risk 
the accusation that they are primarily mathematical artifacts. 

3. Another ground for disagreement lies in the assumption that 
independent factors are the desideratum of any theory of elements. 14 
Not only are all men supposed to have precisely the same basic ele- 
ments in their personalities, but in each life these elements are to be 
regarded as independent of (“freed from the influence of”) all other 

Such an assumption is highly artificial. So interwoven is the fabric 
of personality that it seems almost impossible to think of any patterns 
that are wholly unrelated to others. 15 The picture of the elements of 
personality offered by the theory of “pure” or independent factors, 
and the contrasting picture offered by a theory of focal, but over- 
lapping traits (cf. pp. 326 f.) are suggested in Figure 17. 

4. Finally it should be pointed out that no reconciliation is pos- 
sible between the doctrine of functional autonomy developed in 
Chapter VII and the doctrine of factors. According to the former, 
the course of individuality is one of greater and greater divergence 
from the relatively standard pattern of infancy. The dynamic sub- 
structures of which a personality is composed are unique integrations 
formed in the individual course of experience and heredity. As each 
sub-structure develops it becomes a system of energy sui generis, ob- 
taining, as Tolman has said, a “strangle hold” in its own right. Now 
there may be a few primary strangle holds in the science of wrestling, 
but in the science of personality it is quickly discovered that the 
strangle holds in each life are unique. The factorial approach mis- 

1* Cf. the statement of I. Lorge, “To be useful in psychology a trait must be 
regarded as freed from the influence of other traits.” /. Educ. Psychol., 19 35, 26, 
p. 278. If strict independence is required, then Lorge is correct in acclaiming the 
method of factor analysis, for no other method will reach this objective. 

15 It is said that the assumption of strict independence of factors is not in- 
dispensable to the theory of factor analysis. (J. P. Guilford, Psychometric Methods, 
1936, p. 512. See also by the same author a general defense of factor analysis. Am. J 
Psychol., 2936, 48, 673-683.) Thus far, however, the search has been almost entirely 
for non-correlating, independent “unitary traits” and the chief merit of the method 
is claimed to be precisely this discover)' of non-correlating units. 



(a) The factorial conception of the single personality as a system of independent, 
elements, the elements being the same in different personalities, though varying 
in prominence. 

(b) The trait-conception of a single personality as a system of focal but interde- 
pendent sub-structures, the units being essentially different in every personality 

Figure 17 

Two Contrasting Views of the Nature of the Elements of Personality 



takenly assumes that they are uniform, though varying in weight. 13 

One variation in the factor technique correlates “individuals” 
rather than “qualities” (cf. p. 383). A typological picture results. 
Thus, in a study of esthetic taste certain people are found always to 
prefer softer, pastel tones, and others to prefer saturated, bright 
colors. 17 To find that only certain people have the same taste, the 
same interest, or the same pattern of qualities, is an improvement on 
the contention that all people must be compared on identical vari- 
ables. But “types” are still not individuals, and the method offers no 
real solution to the problem of the neuropsychic elements in any 
single personality. 18 

16 At bottom, of course, this disagreement— as is always the case with conflict- 
ing conceptualizations of data— is due to different intentions . Different theorists have 
quite different ends in view. Speaking in defense of factors Thurstone writes, “The 
final choice of a set of primary reference traits or faculties must be made in terms 
of the discovery that a particular set of reference traits renders most parsimonious 
our comprehension of a great variety of human traits” ( The Vectors of Mind , 
p. 48.) 

But the intent to find the most parsimonious frame of reference under which the 
great variety of human qualities may be ordered, is itself an arbitrary intention. 
From the dynamic point of view it is also an ill-advised goal, at least so it seems 
to R. C. Try on, who writes: 

“No one may object to mathematical- 'actorists employing the rule of par- 
simony in order to simplify their problem. One must realize, however, that the 
factors which come out of the analysis are chosen from many possible sets, and that 
the rule of choice is arbitrary . Now, it happens that the law’ of parsimony is not 
a natural law, but a rule agreed upon among men to simplify their thinking. With 
reference to the psychobiologicai causes of individual differences, nature does not 
appear, however, to work parsimoniously but rather most prodigally. As will be 
shown later, the experimental evidence from psychological and genetics laboratories 
indicates that a very large number of causes determine mental differences. Hence, 
the employment of the rule of parsimony to select out a parsimonious set of fac- 
tors would appear to depict a fiction if such a depiction is urged as a representation 
of psychobiologicai causes.” ( PsychoL Rev., 1935, 42, 427 f.) 

Differing both from Thurstone’s intent to discover a parsimonious set of 
mathematical components and Try on’s intention to discover genetic components 
of endless multiplicity, the desire of the present volume is to discover a type of 
element (trait) that will account without artificiality or undue multiplicity for the 
self-consistency of each individual personality. It is not so much a matter of being 
right or being wrong. It is rather a matter of selecting units that will best represent 
the structure of personality as personality is defined in this volume. 

17 W. Stephenson, Char . & Pers 1935, 4, 17-24. 

18 It is possible that the future will bring a method of factoring the components 
of the single personality considered by itself. In this case a great advance would be 
made, for then it would be possible to determine the cardinal factor for each life 
separately. All the overlap of the focal traits would be consolidated, and the primary 
saturating factor would stand revealed. It would then be possible to tell how much 
of the individual life was influenced by this central factor, and how much is dis- 
sociated from it, and organized around subsidiary factors. 

To accomplish this purpose each individual would have to be considered as a 
“population” and the agreement of his acts with one another would be determined 
in reference to some postulated “intra-individual continuum.” (Cf. F. H. Allporv 
Char. & Pers,, 1937, 5, 202-214.) 



The various factorial techniques undoubtedly have value for cer- 
tain types of problems in psychology and in sociology. But they are 
not qualified to pick out the elements of human personality according 
to the criteria laid down above. Factors fall short of our demand for a 
doctrine of elements that will offer as close an approximation as pos- 
sible to the natural cleavages and individualized structural arrange- 
ments of each single life. The search must go on. 


The doctrine of stimulus-response elements is not necessarily 
nomothetic; it holds that “personality is made up of thousands of in- 
dependent and specific habits.” Whether these habits are regarded as 
common to all individuals is not stipulated; one specificist may think 
that they are, another that they are not. It makes no great difference. 
The point now at issue is different from those previously examined. 
For whether or not specificists think of common or unique habits, 
they all pulverize personality into minute constituent elements. The 
following quotations give the spirit of specificity. 

A widespread misconception of conduct is that it is an expression 
of traits. It is assumed that everyone’s personality is made up of traits 
which are characteristics of the personality. These traits cause one to 
act in certain defined ways. ... It has been found that no one acts 
perfectly consistently with regard to a trait as would be the case if 
conduct was an expression of inner traits . 19 

Over and over, a battery of tests designed to measure traits such as 
persistence, or aggressiveness, or honesty, yields results so unreliable 
and undependable (when compared with other criteria) that one is 
led to question the actual existence of the general traits . 20 

Because of all these complicating factors the study of personality 
becomes extremely difficult, and the analysis of “personality traits” 
must be limited to the particular situation and conditions under which 
they are shown . 21 

A trait is a specific behavior tendency which must be defined in 
terms of a particular stimulus and a particular response . 22 

The specificist contends in brief that the essential element in the 
structure of personality is the habit, and that no organization of a 

10 P. M. Symonds, The Nature of Conduct , 1928, p. 320, by permission of The 
Macmillan Company, publishers. 

20 H. C. Lehmann and P. A. Witty, Am. J. Psychol., 1934, 46, p. 490. 

21 F. L. Goodenough, Developmental Psychology , 1934, p. 444. 

22 F. A. Perrin and IX B. Klein, Psychology , its Methods and Principles , 1926. 
p. 363. 


higher level (such as is suggested in the ordinary usage of the term 
“trait”) exists. 

This doctrine, virtually unknown in the psychological theories of 
other lands, has wide currency in America. Why? One answer is his- 
torical: in this country the reaction against faculty psychology was 
particularly violent. Specifically, it was the teaching of William 
James and E. L. Thorndike that led to a conception of conduct 
largely in terms of habit. Experiments seemed to confirm their teach- 
ing: children had no general powers of the mind that might be edu- 
cated in toto. It was found, for example, that young children could 
not learn cleanliness when taught as an abstract principle, but only 
definite habits of cleanliness, such as brushing the teeth, changing 
soiled clothing, or washing behind the ears. It may be laborious to 
instill habits one at a time, but only in this way— according to the edu- 
cators— can the child learn. (This problem is so complex and far- 
reaching in its implications that it will be treated more fully in the 
next chapter.) From this educational dictum, the habit-psychologists 
found it easy to reach the conclusion that personality is composed of 
“countless specific habits.” The acceptance of specificity was also 
made easier by its congeniality to behaviorism, the theory recently 
prevailing in the American psychological ethos, with its reflex 
pathways and conditioned response, its objectivity, simplicity, and 

Since the doctrine of specificity holds that the individual will do 
in each situation just what he has been trained in that situation to do, 
and nothing else, many American sociologists have likewise favored 
the view, for they too put the weight largely on the situation. If con- 
duct is determined by habits inflexibly bound to the environmental 
stimuli, then the road is clear for the ecological and cultural determi- 
nation of personality. A sociologist wishing to “explain” personality 
in terms of cultural causation has a much easier time with the simple 
theory of habit than with a doctrine of more complex elements in 
which intricate internal and subjective factors are predominant. 

Such are the historical reasons for the doctrine. But its exponents, 
being devotees of the experimental method, require proof of speci- 
ficity. Near at hand they find instances enough that men are not per- 
fectly consistent in their traits. A man may be neat about his person, 
slovenly about his desk; saintly on the Sabbath and diabolic on week- 
days; timid at the office, tyrannical at home. It looks in these cases 
as if it were indeed the situation alone that determined the nature of 



his behavior. The elements are stimulus-bound habits, and not traits. 
But the trouble with this type of casual evidence is that there is much 
of the same order to counterbalance and contradict it. We know 
people who are almost always neat, shy, tactless, cynical, or officious; 
and we predict their behavior correctly in novel situations for which 
there are no specific habits available. 

Hence, the specificists cannot rest their case merely upon cursory 
observation. They invoke experiment. Most frequently they invoke 
the monumental study of the Character Education Inquiry , directed 
by H. Hartshorne and M. A. May. This research is justly famous for 
its ingenious methods, its extensiveness (being probably the largest 
experimental project ever carried out in the field of personality) , for 
its accurate and painstaking treatment of results. The reports com- 
prise three large volumes . 23 By setting before hundreds of children 
concrete tasks in which they had opportunity to react in an individual 
manner, it was possible to study from the records of their behavior 
the evidence for and against the existence of such alleged traits as 
deception, help-fulness, co-operativeness, persistence, and self-control. 
The final conclusion was that such qualities as these are “groups of 
specific habits rather than general traits.” 24 This imposing investiga- 
tion is so influential, and so often cited as evidence for the specificist’s 
position, that its results and interpretations must be examined with 

1. Our first discovery is that the low correlations found between 
the tests employed prove only that children are not consistent in the 
scnne way, not that they are inconsistent with themselves. This is an 
exceedingly important discovery. 

In studying dishonesty, for example, children were presented with 
tasks giving them repeated opportunities to be deceitful— to steal pen- 
nies, to correct their school papers to their own advantage, to cheat at 
games, to lie about their cheating. When the records for the entire 
population of children were studied it was found that there was little 
tendency for the youngsters to be uniformly honest or dishonest in 
all types of behavior. For example, the correlation between the score 

23 H. Hartshorne and M. A. May, Vol. I, Studies in Deceit , 1928; Vol. II, 
Studies in Service and Self-Control , 1929; Vol. Ill (with F. K. Shuttle worth) , Studies 
in the Organization of Character , 1930. A useful secondary account of the methods 
and findings is contained in P. M. Symonds, Diagnosing Personality and Conduct* 
1932, chap. ix. 

Op. cit., Vol. Ill, p. 1. 


for stealing pennies (which by itself turned out to be a fairly con- 
sistent habit) and the scores for telling lies about their cheating 
(which by itself was another rather consistent habit), was only 
+.132. Clearly, the dishonest habit aroused by one of these situations 
is quite independent of the dishonest habit in the other. The children 
who steal do not necessarily lie. There is no common trait existing 
in all children in the same way. But, on the other hand, the habit of 

(Dotted ellipse represents the trait as conceived by investigators in the Charac- 
ter Education Inquiry, solid ellipses, possible personal traits overlooked by them.) 

stealing pennies may, in each child’s life, be an integral part of some 
trait, even though it is rarely related to the habit of lying. Figure 18 
shows the possibility. 

It may be that child A steals pennies because he has a consistent 
personal trait of bravado based upon his admiration for the gangsters 
he reads about in the tabloids and sees on the screen; child B steals 
because he has a persistent interest in tools and mechanics that drives 
him to buy more equipment than he can honestly afford; child C, suf- 
fering from a gnawing feeling of social inferiority, steals pennies to 
purchase candy to buy his way into favor with his playmates. Child D 
does not steal pennies, but he lies about his cheating, not because he 
has a general trait of dishonesty, but because he has a general trait 
of timidity (fear of consequences); child E lies because he is afraid of 
hurting the feelings of the teacher whom he adores; child F lies be- 
cause he is greedy for praise. Each of these children behaved as he did 



toward these tests, not because he had specific habits, but because he 
had some deep-lying and characteristic trait. All that the C.E.L dis- 
covered was that the particular trait of honesty as defined in the usual 
ethical terms and tested in various conventional situations, was not 
one of which the children possessed constant individual degrees, espe- 
cially in the face of perhaps a stronger tendency of each child to ex- 
press some trait other than honesty through the behavior of lying and 
stealing. The children did not all have the same trait, but they had 
nevertheless their own traits. 

2. The investigators based their research upon social and ethical 
concepts . The methods used were not devised from the point of view 
of child psychology, but from the point of view of society and its 
conduct values. Our culture places a premium upon honesty, service, 
and self-control, but these items in the social code seldom correspond 
precisely to the form of mental organization found in adults, and still 
less to the unsocialized dispositions of childhood. In terms of the dis- 
tinction drawn in Chapter II, the investigators confused their research 
at the outset by selecting characterial rubrics for their starting point. 
A study of good qualities and bad qualities is not the same as a study 
of natural qualities. The study of personality is difficult enough 
without complicating it at the beginning with ethical evaluation. 25 

3. Whenever moral standards are involved, the question of the 
age of the subjects is of greatest importance. Older children and 
adults learn gradually the requirements of social custom; they come 
to know what is meant by honesty in their culture, by service, and by 
self-control. What is more, they may introcept these prevalent ideals 
into their own lives and guided by these standards, may develop in- 
tegrated dispositions roughly corresponding to the ideals. This is the 
process of socialization. According to the principle of functional 
autonomy such acquired traits may in time become exceedingly 
dynamic, causing sharp pangs of conscience whenever their dictates 
are violated. Rut such socialized traits should not be expected in the 
younger child. Much of the evidence of the C.E.L indeed demon- 
strates their gradual development with age. The older child more 
frequently than the younger guides his conduct in accordance with 
the social ideal. 

25 It should be pointed out that the investigators obtained their most consistent 
results with the trait Gf persistence , which, for example, showed the greatest pro- 
gressive integration with age. This trait is the only one in the schedule that seems 
entirely free from extraneous definition in terms of social code; it is an authentic 
psychological conception. 


The evidence shows likewise that the children grow more and 
more consistent in respect to the positive social ideals, but not in re- 
spect to vices or anti-social conduct. 20 In other words, the pressure 
of the environment leads gradually to conformity with the social 
code, and the conformity is flexible and generalized. Only wrong- 
doing is specific. This result is just what one would expect in the 
course of normal socialization when emphasis in training is placed 
upon virtuous ideals, and only occasional lapses are allowed. 

4. Some arbitrariness is always involved in the interpretation of 
complex statistical results. Surveying the myriads of intercorrelations 
between the children’s scores on many tests, one is struck by the pre- 
vailingly positive association that is revealed. The coefficients are low, 
to be sure, but even so, why should they be positive? Considering the 
various insufficiencies of the methods used, and the preoccupation 
with conduct common to a whole population of children, to the ex- 
clusion of conduct characteristic of each child, it is surprising that the 
results were even slightly positive. What do these low positive corre- 
lations mean? Some investigators say specificity, others generality. No 
one knows. Hartshorne and May have chosen “to follow the evi- 
dences of specificity to their logical conclusion.” 27 This, in the face 
of the fact (reported on the same page) that “the twenty-three tests 
used in securing our total character score, for example, intercorrelate 
+.30 on the average.” Instead of deducing specificity from this 
matrix, Mailer, an associate in the Inquiry, finds it adequate evidence 
for postulating a u c” factor of character which is present in all be- 
havior, saturating it with general quality and common strength. 28 The 
hypothesis of a general factor of character (derived from the same 
data!) is of course the complete antithesis of the doctrine of speci- 
ficity. Here is a pointed illustration of the fact that correlational 
methods per se solve no problems, simply because all coefficients (o 
and 1. 00 perhaps excepted) are intrinsically ambiguous and need 

5. Finally, whether specificity or generality is found in the struc- 
ture of personality depends to a large extent, not only upon the inter- 
pretation of quantitative evidence, but upon the methods used. Em- 
ploying large populations of children and myriad tests Is a very dif- 

26 Op. cit. 9 Vol. Ill, p. 375. 

Op. cit Vol. Ill, p. 364. 

28J. B. Mailer, /. Soc. Psychol . , 1934, 3, 97*101. C is defined as the readiness to 
forego immediate gain for the sake of remote but greater gain. 


toward these tests, not because he had specific habits, but because he 
had some deep-lying and characteristic trait. All that the C.E.I. dis- 
covered was that the particular trait of honesty as defined in the usual 
ethical terms and tested in various conventional situations, was not 
one of which the children possessed constant individual degrees, espe- 
cially in the face of perhaps a stronger tendency of each child to ex- 
press some trait other than honesty through the behavior of lying and 
stealing. The children did not all have the same trait, but they had 
nevertheless their own traits. 

2. The investigators based their research upon social and ethical 
concepts. The methods used were not devised from the point of view 
of child psychology, but from the point of view of society and its 
conduct values. Our culture places a premium upon honesty, service, 
and self-control, but these items in the social code seldom correspond 
precisely to the form of mental organization found in adults, and still 
less to the unsocialized dispositions of childhood. In terms of the dis- 
tinction drawn in Chapter II, the investigators confused their research 
at the outset by selecting characterial rubrics for their starting point. 
A study of good qualities and bad qualities is not the same as a study 
of natural qualities. The study of personality is difficult enough 
without complicating it at the beginning with ethical evaluation. 25 

3. Whenever moral standards are involved, the question of the 
age of the subjects is of greatest importance. Older children and 
adults learn gradually the requirements of social custom; they come 
to know what is meant by honesty in their culture, by service, and by 
self-control. What is more, they may introcept these prevalent ideals 
into their own lives and guided by these standards, may develop in- 
tegrated dispositions roughly corresponding to the ideals. This is the 
process of socialization. According to the principle of functional 
autonomy such acquired traits may in time become exceedingly 
dynamic, causing sharp pangs of conscience whenever their dictates 
are violated. But such socialized traits should not be expected in the 
younger child. Much of the evidence of the C.E.I. indeed demon- 
strates their gradual development with age. The older child more 
frequently than the younger guides his conduct in accordance with 
the social ideal. 

2!i It should be pointed out that the investigators obtained their most consistent 
results with the trait of persistence, which, for example, showed the greatest pro- 
gressive integration with age. This trait is the only one in the schedule that seems 
entirely free from extraneous definition in terms of social code; it is an authentic 
psychological conception. 


The evidence shows likewise that the children grow more and 
more consistent in respect to the positive social ideals, but not in re- 
spect to vices or anti-social conduct. 21 '’ In other words, the pressure 
of the environment leads gradually to conformity with the social 
code, and the conformity is flexible and generalized. Only wrong- 
doing is specific. This result is just what one would expect in the 
course of normal socialization when emphasis in training is placed 
upon virtuous ideals, and only occasional lapses are allowed. 

4. Some arbitrariness is always involved in the interpretation of 
complex statistical results. Surveying the myriads of intercorrelations 
between the children’s scores on many tests, one is struck by the pre- 
vailingly positive association that is revealed. The coefficients are low, 
to be sure, but even so, why should they be positive? Considering the 
various insufficiencies of the methods used, and the preoccupation 
with conduct common to a whole population of children, to the ex- 
clusion of conduct characteristic of each child, it is surprising that the 
results were even slightly positive. What do these low positive corre- 
lations mean? Some investigators say specificity, others generality. No 
one knows. Hartshorne and May have chosen “to follow the evi- 
dences of specificity to their logical conclusion.” 27 This, in the face 
of the fact (reported on the same page) that “the twenty-three tests 
used in securing our total character score, for example, intercorrelate 
+.30 on the average.” Instead of deducing specificity from this 
matrix, Mailer, an associate in the Inquiry, finds it adequate evidence 
for postulating a “c” factor of character which is present in all be- 
havior, saturating it with general quality and common strength. 28 The 
hypothesis of a general factor of character (derived from the same 
data!) is of course the complete antithesis of the doctrine of speci- 
ficity. Here is a pointed illustration of the fact that correlational 
methods per se solve no problems, simply because all coefficients (o 
and x.oo perhaps excepted) are intrinsically ambiguous and need 

5. Finally, whether specificity or generality is found in the struc- 
ture of personality depends to a large extent, not only upon the inter- 
pretation of quantitative evidence, but upon the methods used. Em- 
ploying large populations of children and myriad tests is a very dif- 

26 Op. cit Vol. III, p. 375. 

27 Op. cit., Vol. Ill, p. 364. 

23 J. B. Mailer, /. Soc. Psychol, 1934, 5, 97-101. C is defined as the readiness to 
forego immediate gain for the sake of remote but greater gain. 



ferent procedure from studying more intensively the behavior of 
fewer subjects at a more mature age. It is astonishing what different 
results this latter method brings when applied to the same problem of 
the self-consistency of the traits of honesty and deceit . 

At the Harvard Psychological Clinic, D. W. MacKinnon set for his 
93 adult subjects a series of difficult problems whose solutions were 
to be worked out in solitude during an hour’s experimental session. 29 
The answers to these problems lay before the subject in pamphlets, 
and certain answers the subject was permitted to consult if he chose; 
the remainder of the answers were forbidden to him. Through a one- 
way screen the experimenter observed each instance where the sub- 
jects violated the prohibition. Approximately half of the subjects were 
violators and half non-violators. 

The first evidence of trait-consistency comes from the experi- 
menter’s predictions, made after approximately a five minute inter- 
view with each subject. On the basis of first impressions he prophesied 
which subjects would violate the prohibition and which would not. 
This method was employed with 74 of the subjects. The prophecy 
proved to be correct in 69 per cent of the cases. The success was 
greater in the case of non-violators (a fact that suggests the finding 
of Hartshome and May that integration— or spread of a trait— is 
greater for the honest than for the dishonest subjects). 

Of the 34 violators, correct predictions were made for only 62 per 
cent. What of those violators for whom false predictions were made? 
The error, it appeared, was due chiefly to the fact that the unexpected 
violators were also atypical violators. That is to say, they were the 
ones who later showed unmistakable signs of guilt (not shown by 
typical violators). Excepting for the deed itself the atypical violators 
behaved like the non-violators. They were, it seems, really honest 
people, who under the stress of the experimental situation had yielded 
to the temptation, violating their own customary standards. The very 
fact that they showed guilt, remorse, and repression following the 
violation suggests that in general they had a dependable trait of 

The typical violators, on the other hand, were consistent with them- 
selves. They lied to the experimenter more frequently than did the 
non-violators or the atypical violators. They lied even about the an- 
swers they were permitted to consult! These subjects also denied that 
they were ordinarily troubled by guilt (only 29 per cent admitted 
feelings of guilt in everyday life as compared with 75 per cent of the 

29 D. W. MacKinnon, “The Violation of Prohibitions in the Solving of Prob« 
lems,” to be published in Explorations in Personality (H. A. Murray, et al.). 


2 55 

non-violators). They maintained that they would not feel guilty “even 
if they had cheated! ” 

The honest subjects also presented a consistent picture. They were 
not only recognized as such (in the majority of cases) on first ac- 
quaintance, but they did not lie about consulting the permitted an- 
swers. They also showed little tendency to blame the experimenter 
for giving them such hard problems, or to be unduly aggressive (both 
of which characteristics were found in the violators). They admitted 
suffering from guilt whenever in everyday life they violate their own 
standards of conduct. 

Such results entirely contradict the hypothesis of specificity. 
There are honest people, dishonest people, and atypical people- 
honest in most respects, but not always capable of resisting tempta- 
tion. An intensive study of this sort, then, leads to the discovery in 
most instances of generality in the trait of trustworthiness; whereas 
the extensive (statistical) study of the C.E.I. gives the honors to 
specificity! Murphy and Jensen are perhaps right when they say, 
“Honesty is either a general characteristic or a set of specific habits, 
depending upon your interest and your emphasis.” 30 They might 
have added, “depending also upon your method, and upon the par- 
ticular individual you happen to be studying.” 


The doctrine that all behavior (and therefore personality) is com- 
posed entirely of countless specific habits is, for several reasons, un- 

i. In the first place, the chief evidence adduced for it has come 
from procedures that may be incapable of discovering higher units 
of organization in personality. The study of moral qualities is a case 
in point. Are they the place to look for maximum personal integra- 
tion? With advancing socialization, to be sure, the psychological 
make-up of a person may become centered in such a way as to corre- 
spond in some degree with social norms, but transgressions from these 
norms do not prove that all conduct is specific. Such transgressions 
may occur because of profoundly consistent dispositions, more funda- 
mental to the person than the conventional codes. Example: 

A boy of ten, accustomed to the respect of his own playmates at 
home, found himself for the summer among older boys who looked 
3 ® G. Murphy and F. Jensen, Approaches to Personality, 1932, p. 385. 


2 5 (5 

down upon him as an outsider and as a punk. He felt frustrated and 
chagrined. One day a member of the gang proposed “swiping” a few 
bars of candy from the corner store. At first the lad’s habits of 
honesty prompted him to resist the suggestion, but when the gang 
ridiculed him, his major desire for admiration and social standing be- 
came aroused. Alone he undertook the larceny, and in a few moments 
emerged from the store with a plentiful supply of chocolate bars. The 
less stable habits of honesty were destroyed by a stronger and more 
organized trait. The boy was consistent enough with himself, but his 
consistency did not happen to correspond to the social ideal. 

The error of probing for consistency in the wrong place (and 
failing to find it, pronouncing in favor of specificity) has been likened 
by G. B. Watson to the absurdity of asking whether a person using 
the public library has a trait causing him to take out only books with 
red or with blue covers. Of course he hasn’t. If only the bindings 
were studied, no consistency should be expected. But if the subject- 
matter of the chosen books was investigated, well organized traits of 
interest would appear. 31 Unless consistency is sought in the right 
direction, consistency will not be found. 

2. A related error lies in employing exclusively the results of 
mass investigation. Statistical methods are ordinarily applied only to 
those variables to which all people may be ordered. If many people 
do not happen to fit the variable then the illusion of specificity re- 
sults. Low correlations between habits of behavior (as demonstrated 
in Figure iS) mean at the most that different individuals are not all 
consistent in the same way; they never prove that the individual is 
not consistent with himself in his own way. 

3. The danger of resting the theory of specificity upon studies 
of young children has been mentioned. Children show less integra- 
tion, as a rule, than adults. The evidence presented by specificists 
usually shows, in fact, that older children are more consistent than 
younger children. 

4. The evidence from “personality tests” is by no means as favor- 
able to the doctrine of specificity as some writers imply (cf. quota- 
tion from Lehmann and Witty, p. 248). If a scale possesses reliability 
(internal consistency) the fact can signify only one thing, namely, 
that people respond to related diagnostic items in a way characteristic 
of themselves; their response shows that they are in varying degrees 
habitually ascendant , extroverted , persevering, or radical. Reliability 

31 G. B. Watson, Char. e> Pm\, 1933, 2, p. 69, 


of a many-itemed scale is pr'rma facie evidence for some kind of gen- 
erality in conduct. 

This fact is all the more arresting when one considers that a 
standardized scale, in order to achieve reliability, requires that all the 
subjects be self-consistent in respect to the same common variable. 
Figure 18 showed that consistency in respect to a common variable 
may be negligible without proving the case for specificity, since each 
individual has his own peculiar traits with which he may be self- 
consistent. But when, as is the case with these scales, high reliability 
is found in common variables as well, the evidence for the existence 
of complex units of organization mounts even higher. 

5. If we become intentionally naive and ask whether any spec- 
ificist in daily life can believe and be guided by his theory, we dis- 
cover swiftly that he cannot. To be in keeping with his own doc- 
trine he would never be entitled to apply descriptive adjectives to 
persons. He could not say that his friend is trustworthy, affable, or 
humorous; that his child is high-strung, dominant, or sociable; that his 
wife is kind or tactful. In the midst of an argument for specificity, 
in a passage commenting on the filling out of questionnaires, one 
author makes a fatally contradictory observation, “The truthful person 
finds it very difficult to be untruthful”— a statement, of course, not 
of specific habits but of traits! No one can for long live or think 
according to the sparse concept of specificity. 

6. Everything that is known concerning the integrative action 
of the central nervous system is incompatible with the doctrine of 
specificity. Belief in the “neural groove,” wherein a specific habit was 
once supposed to reside, has been abandoned. Though little enough 
is known concerning the neural equivalents of generalized disposi- 
tions, the trend of evidence so far as it goes favors such a theory 
rather than one of insulated habits. 


Thus far our search for basic psychological components of per- 
sonality has not seemed very rewarding. A number of schemes have 
been examined and, though each has its merits for certain purposes, 
none serves the end we here have in view. None is concerned with 
the strictly individualized units of personality that are the true carriers 
of individuality in behavior. For the most part they are nomo- 
thetically conceived, having as their aim the facilitating of compari- 


sons between one person and all others in respect to some one com- 
mon aspect of personality. Other schemes that are not necessarily 
nomothetic are for the most part specificistic in conception or highly 
miscellaneous and of no theoretical value. At any rate none of the 
schemes considered in this chapter defines the level of complex but 
well-structured personal dispositions that can serve as the most suit- 
able tool of analysis. 

More promising is the doctrine of sentiments, discussed on p. 89, 
where individuality of affective organization is better allowed. Help- 
ful too are certain conceptions drawn from Gestalt Psychology, such, 
for example, as Wertheimer’s radix, Koffka’s ego-systems, Lewin’s 
regions , tension-systems, inner personal strata, and the like. In one 
way or another these various conceptions suggest improved units 
of analysis. The special merits of some of these proposals will be con- 
sidered later and their applications discussed. In the meantime one 
more critical duty lies ahead, before the ground is completely cleared 
for a constructive theory of traits. 



The problem of the organization and structure of personality is 
not separate from the more inclusive problem of mental organization, 
but is merely one phase of it. The only real difference is that when- 
ever interest centers in personality, the emphasis is on the lasting or 
permanent characteristics of mental organization; whereas in the older 
association psychology, and in the modem study of perception, learn- 
ing and thinking, it is a question of the organization of mental content 
into somewhat more transient sequences and constellations. But if one 
could tell precisely how temporary arrangements come about in the 
human mind, one would probably have the key to the more perma- 
nent arrangements, and vice versa. At all events, in the present stage 
of psychological knowledge, it is impossible to pursue the problem 
of the organization of personality apart from the more inclusive prob- 
lem of mental organization. As the years pass this problem is less 
frequently formulated in the older terms of the “laws of association,” 
but whatever guise it takes, the issue is still the same, and it is the 
pivotal issue of all psychological science. 

The most prevalent doctrine of mental organization in American 
psychological theory is the doctrine of identical elements, or, as it is 
often called, of “partial identity.” According to it the consistency of 
an individual’s conduct is due to his ability to discern minute elements 
or aspects of an environmental situation that are identical with ele- 
ments or aspects of situations encountered previously, and to make a 
corresponding response wherever these elements occur. To this so- 
called “ transfer ” in conduct Symonds has given the name “confact” 
(a term corresponding to “concept”— which, according to the doc- 
trine of partial identity, arises likewise through the ability of mind to 
detect minute perceptual identities) . Both concepts and confacts “are 
responses to a single element which may be common to a number 
of otherwise varying situations. ... A concept is a mental or verbal 
response, whereas a confact is a conduct response.” 1 

1 P. M. Symonds, The Nature of Conduct, 1928, p. 167. 

2 6o identical elements 

Truthfulness, courage, loyalty, frugality, generosity, kindliness, 
and other alleged traits are to be explained by the capacity of an 
individual to respond with the same sequence of habits to identical 
elements in otherwise varying situations. For example, a boy must 
never be thought to develop a trait of courtesy. Rather he “learns 
to take off a specific cap when coming in a specific door and in the 
presence of his mother. But in time he may take off his cap or hat 
or whatever he has on his head when entering any door, in any 
house whatsoever, whether or not in the presence of a person.” 2 
Courtesy, then, is nothing more than the repetition of the same habits 
over and over again when provoked by stimuli previously associated 
with these habits, or when there is partial identity in different 
stimulus fields, or when the habits themselves are related by means 
of some common bond. Such a tenuous principle of organization 
naturally leads to the conclusion that “generalization of conduct does 
not extend as far as most people suppose, with the result that conduct 
for the large masses of people remains unorganized, a rather loose 
bundle of unrelated and dissociated habits.” 3 

A diagram will make the theory clear. The boy in the illustration 
has learned, by dint of drill, that when he enters an outside door (a) 
in his own home (Stimulus Field I), he should take off his cap 
(Habit i), wipe his shoes (Habit 2), greet the occupants (Habit 3). 
Now, entrance doors have something in common (identical elements 
in different stimulus fields) : they all provoke the boy to perform the 
same three habits (as, for example, when he enters a neighbor’s door, 
d, Stimulus Field II. A misinformed observer might remark that here 
is a courteous lad. Not at all; he has no general trait; his three habits 
are automatically set into motion whenever he encounters identical 
elements in different fields. 

Now, the habits themselves are also capable of having identical 
features, so that the arousal of one may arouse another (by virtue of 
redintegrative action, cf. p. 203). Thus, greeting the occupants of 
the house (Habit 3) might through some commonly associated verbal 
components— for example, “Hello, how are you?”— set in motion a 
further inquiry concerning their health (Habit 4), which in turn by 
virtue perhaps of some commonly associated facial expression, might 
suggest a friendly smile (Habit 5). Finally a friendly smile through 

2 P. M. Symonds, The Nature of Conduct, 1928, p» 294. 

3 Ibid., p. 325. 


further specific association leads to some additional act of “courtesy” 
such as asking his friend out to play (Habit 6). 

Thus it is that identical elements in the stimulus fields, or in the 
neuropsychic structure, are responsible for the quasi-consistency of 
the boy’s conduct. What is denied is that any generalized disposition 

Identity may reside either in corresponding elements of the stimulus fields, or 
else in some neuropsychic components of the habits. Solid connecting lines (?) indi- 
cate such identities. Arrows show lines of stimulation, and the dotted ellipse repre- 
sents the hypothetical trait of “courtesy” denied by the theory. 

exists to which the name “courtesy” might be properly attached. 
The boy is basically a creature of specific habits. Whatever relation 
there may be between the six habits here enumerated is due entirely 
to some still more minute and specific components of these habits, in 
short, to some identical elements. 

This, then, is the picture of the structure of personality accord- 
ing to the theory of partial identity. It is a serious question whether 
the terms “generalization” and “integration” can be applied at all to 
such a thread-like connectionism. It is an even more serious question 



whether the inward cohesion in the structure of personality can be 
accounted for in this way. 


Any law of learning is at the same time a law of the development 
of personality (cf. p. 151). Now, one of the outstanding problems in 
the psychology of learning is transfer of training; and, as might be 
expected, it turns out that this same problem is equally important for 
the theoretical psychology of personality. 

Suppose it should come about— as the theory of identical elements 
avers— that 

Training the mind means the development of thousands of particu- 
lar independent capacities, the formation of countless particular habits, 
for the working of any mental capacity depends upon the concrete 
data with which it works. Improvement of any one mental function 
or activity will improve others only in so far as they possess elements 
■; ; common to it also. The amount of identical elements in different men- 
tal functions and the amount of general influence from special training 
are much less than common opinion supposes . 4 

Suppose such is the case. What then is the picture of the developing 
personality? It is one in which specific adjustments to specific stimuli 
are the outstanding fact— the recurrence of the same movement in the 
face of stimulus situations that are partially the same; sequences of 
closely related habits would be set into operation; throughout the 
process a marked rigidity would prevail. In training a child it would 
be of no avail to teach him general principles and abstract ideals, for 
his models of conduct can never be anything but specific. 5 Novel 
situations would leave the child bewildered and impotent, or throw 
him back upon “instinct,” unless some familiar item in the novel 
situation rescued him from the dilemma by reinstating an available 
habit, or habit sequence. The picture is one of invariance of the 
stimulus; invariance of the mental set, and invariance of the response. 6 

4 E. L. Thorndike, Principles of Teaching , Seiler, 1906, p. 248. 

5 The modem movement of “character education” has been greatly influenced 
by this belief. In its literature one reads remarkable dicta, for example, to the effect 
that no child can be taught virtue except “by doing specific things in specific situa- 
tions.” As an example of this literature, see H. Hartshorne, Character in Human 
Relations , 1932. 

6 But usually a habit that is reinstated invariandy appears merely inappropriate 
and ridiculous. There is a familiar story about an old soldier coming from the viJ- 


Suppose this theory of learning is false. Suppose it is only the 
young child or the mentally undeveloped who must be taught in 
terms of habits, and taught in this way only because his capacity 
for generalization has not matured. Suppose that with time these 
specific habits can become authentically generalized, and thus re- 
placed by much more fluid and flexible dynamic dispositions, of the 
order of traits. Suppose that the maturing individual represents such 
traits to himself subjectively as ideals of conduct. A youth, for in- 
stance, accepting courtesy as an ideal worthy to guide his conduct, 
would not thereby be bound by six habits nor sixty, nor only by such 
habits as have among themselves partial identities. If suddenly re- 
moved to China, where courtesy requires the inversion of most of 
his habits, he would readily enough readjust them and behave in a 
fashion that would reverse his specific habits and yet be consistent 
with his general ideal. Habits would not be a ball and chain, but 
would readily atrophy or even reverse themselves, whenever the 
stable and generalized dispositions of personality demanded. 7 

Returning to the problem of transfer, psychology, we may say, 
has passed through two distinct periods in reference to this important 
question, and is now slowly emerging into a third. The first period 

lage store. A wag shouted the command Attention! and the old soldier leaped to 
position, dropping his basket of eggs. In every particular the old man was mani- 
festing transfer through the medium of identical elements. The command was the ele- 
ment common to two life settings; of itself it reinstated a previously formed habit. 
But such behavior is as comic as it is rare. (Bergson based a whole theory of laughter 
upon such “mechanical inelasticity .”) 

7 Some time ago habit training was introduced into prisons on the assumption 
that good habits made good citizens. The prisoners were taught hygienic habits, 
vocational habits, and were paid wages that they were forced to share regularly 
with their families. During a long period of incarceration the prisoners received an 
impressive amount of such habit training, which was designed in such a way that 
identical elements in the world outside the prison should cause these salutary habits 
to function when the prisoner was freed. 

The policy did not work. The study of S. and E. T. Glueck {$oo Criminal 
Careers , 1930) shows that 80 per cent of the men discharged from a reformatory 
where many approved methods of habit training were in use, were not reformed 
five to fifteen years later, but continued in their course of crime. The gradual train- 
ing of habits and reconditioning, with reliance upon identical elements within and 
without the prison walls to effect the transfer, produced few reformations, if any 
at all. To a prisoner with a dominant anti-social outlook, or other antagonistic 
traits, such habit training is worthless, for he is in no frame of mind to put it to 
use. On the other hand, in those rare cases where the dominating attitude and goals 
are altered, habit training is of secondary importance. The reformed prisoner will 
find ways of learning to live more hygienically, of taking care of his family, of 
fulfilling his responsibilities, provided only that his interests and ideals are altered 
And they are not altered by mere routine drill. 



was characterized by blind faith in limitless transfer, wherein it was 
assumed that any type of training improved mental power in general, 
or at least the power of some broad faculty of the mind, such as 
reasoning ability or memory. Even today, in less “progressive” 
schools, teachers believe, and assure their students, that the study of 
geometry or Latin, though unsavory in itself, is valuable because 
it automatically trains “logical power,” or “memory,” or perhaps 
“will-power.” 8 

It was Thorndike who, following the suspicions of James, Hins- 
dale, and others, 9 made a devastating attack upon this widespread 
theory of “formal discipline,” and instituted the second epoch. In 
1901 he published, in collaboration with R. S. Woodworth, one of 
the earliest experimental investigations of the problem. 10 This study, 
according to Gates, 

resulted in a complete overthrow of the older educational theory of 
formal discipline. In place of this doctrine was offered the theory of 
transfer of training, which, in brief, states that improvement in think- 
ing, reasoning, neatness, honesty, and the like, is to be found in the 
development of innumerable particular habits, and that these habits are 
likely to remain imbedded in the situation in which they are devel- 
oped. A corollary to the theory is that such habits transfer from a 
situation in which they were developed to other situations roughly 
in proportion to the degree to which the two settings have elements 
in common. 11 

Thorndike himself saw the following educational consequences 
in this early experimental work. 

By doubling a boy’s reasoning power in arithmetical problems we 
do not double it for formal grammar or chess or economic history 01 
theories of evolution. By tripling the accuracy of movement in fin- 
gering exercises we do not triple it for typewriting, playing billiards 
or painting. The gain of courage in the game of football is never 

8 Instances of the older pedagogical literature of this type are C. Aiken, Methods 
for Mind Training, 1889, and J. Payot, The Education of the Will , trans. 1909. 

9 Cf. W. James, Principles of Psychology , 1890, I, 666-668; B. A. Hinsdale, 
“The Dogma of Formal Discipline,” Proceed. N. E. A., 1894. 

i° Psychol. Rev., 1901, 8, 247-261, 384-395, 553-564. All citations from the work 
of E. L. Thorndike in this chapter are perforce from his earlier publications. In 
neither Human Learning , 1931, nor The Fundamentals of Learning , 1932, is there an 
explicit defense (or repudiation) of the theory of identical elements. 

11 A, I. Gates, quotation from chap, iii of Psychology at Work (ed. by P. S. 
Achilles) , McGraw-Hill Book Company, 1932. 


equaled by the gain in moral courage or resistance to intellectual ob- 
stacles. The real question is not, “Does improvement of one function 
alter others?” but, “to what extent, and how, does it?” 

The answer which I shall try to defend is that a change in one func- 
tion alters any other only in so far as the two functions have as factors 
common elements. ... To take a concrete example, improvement in 
addition will alter one’s ability in multiplication because addition is 
absolutely identical with a part of multiplication, and because certain 
other processes— e.g., eye movements and the inhibition of all save 
arithmetical impulses— are in part common to the two functions. 12 

Thorndike’s influential theory may be summarized as follows. 
Each mental operation is by nature a distinct process, beginning with 
a stimulus that traces a neural path until a motor adjustment is 
achieved. The neural path once traversed is now in a state of reduced 
resistance, and will function more readily when stimulated again. 
The various stimulus-situations of life frequently have features in 
common, so that in what seems to be a new situation, there are really 
elements present that have been present before. This duplication of 
features in two stimulus-situations is responsible for whatever im- 
provement is manifested in the second response. The second response 
will be made more quickly or more accurately, that is, will show 
transfer-effect in proportion to the amount of overlap. 

In the case of arithmetic, practicing addition should show a 
greater amount of transfer to the learning of multiplication than to 
the learning of division, for the reason that the first two processes 
have more elements in common ( c , d, e, and f, in Figure 20) ; whereas 
addition and division have fewer factors in common (only e and f). 

One weakness of the diagram is immediately apparent, namely, 
that the “elements” constituting each mathematical operation are by 
no means elementary (though derived from Thorndike’s own analysis 
of arithmetical operations) . But, as will later be shown, this dubiety 
concerning the nature of elements is inherent in the theory itself. 
For the moment all that is required is to grasp the intent of this 
theory, and to appreciate its influence upon doctrines concerning the 
structure of personality. As instances of this influence the following 
quotations will serve. 

Character traits are not general traits available to apply to all sorts of 
situations— but rather the reverse— a specific thing in each situation. . . . 
Psychology no longer recognizes honesty, e.g., as a general trait which 

12 Educational Psychology, Teachers College, 1903, pp. 80 f. 


once “acquired” would make one honest in whatever he did— rather is 
it true that we come to be regarded as honest by the way we respond 
to individual situations. . . . Our interest has shifted from “being 
good” to “doing good,” from “being honest” as an intangible abstrac- 

r a. mental set to add 
! b. attending to long vertical column 



^ J c. table of additions 

d. process of carrying numbers 

e. set to inhibit non-mathematical ideas 
b f. tenseness required for accuracy 

g. mental set to multiply 

h. multiplication tables 

i. treatment of zeros in multiplier 

j. treatment of zeros in multiplicand 



' a. mental set to add 

b. attending to long vertical column 

c. table of additions 

< d. process of carrying numbers 

e. set to inhibit non-mathematical ideas 
L f. tenseness required for accuracy 

•i w. subtraction tables 

x. borrowing 

y. constant reference to divisor 

z. mental set to divide 

Figure 20 

Some Hypothetical Elements Involved in Different 
Arithmetical Operations 

tion to “doing honestly” in the concrete situations, or from “being 
neat” to “doing neatly” in action . 13 

We all know that individuals may be courteous, kind and generous 
in company or in business relations, and at the same time be rude, 
cruel, selfish at home. An unscrupulous business man may be the very 
essence of honesty in his relations with his friends, his club, his 
church. The kind doting father may be an employer of children 

18 From a memorandum issued by one of the leading “character forming” agen- 
cies in America. 


under sweat-shop conditions. So kindness, honesty, generosity are not 

traits that transfer from one set of circumstances to all others. 14 


A problem so central in educational theory and practice has 
naturally been submitted time and again to experimental investiga- 
tion. Since the classic experiment of William James (1890) there have 
been upwards of three hundred experimental reports devoted to the 
subject. James was interested in the possibility of training memory. 
Putting it to test, he memorized 158 lines of Hugo’s Satyr, and then 
for practice, memorized the entire first book of Paradise Lost; After 
this intensive training he returned to Hugo’s poem; another 158 lines 
required 15 1% minutes, as compared with the 132 minutes required 
for the first 158 lines. There was manifestly no transfer, but James 
doubted the results since he was fatigued at the time of the second 
test. Later repetitions of the experiment fully justified his doubts. 15 

The customary procedure in all modern experiments on transfer 
is to compare the gain for a practiced group of subjects with the gain 
(if any) for a control group, the latter being in every respect equal 
to the experimental group, and like it, tested at the beginning and 
end of the experiment, without, however, any specific training in the 
interval. The procedure is: 

Training Group: test in Function 1; training in F 2; retest in F 1. 
Control Group: test in Function 1; no training in F 2; retest in F 1. 

When the numerous published experiments are compared with 
one another little agreement is found in their results, no doubt be- 
cause of the widely different conditions under which the experi- 
ments were conducted, the different types of material they employed, 
and the different mental functions studied. In spite of this disagree- 
ment, there are few dissenting voices to the conclusion that transfer 
does take place, in varying amounts, depending on circumstances. 
Transfer fails only when some extraneous condition (such as James’s 
fatigue) precludes it, or when the tested and the trained functions 
stand in some kind of antithetical relation to one another, bringing 
about interference instead of the expected aid; the latter condition 
is known as “negative transfer.” The experimenters’ explanations of 

14 R. Pintner, Educational Psychology , 1929, p. 2 65. 

15 H. A. Peterson, Psychol. Rev., 1912, 19, p. 491. 


their results are almost as numerous as the investigations themselves. 
Orata presents a summary of these interpretations, most of which in 
one way or another subscribe to the theory of identical elements, 
though some totally reject it. 1 ® 

A detailed summary would serve no useful purpose here. Most 
of the experiments suffer from two serious weaknesses. For one thing, 
they often employ absurdly inconsequential tasks for training. To 
cross out p’s and q’s in a line of type, or to draw several hundred 
senseless designs, is a poor basis for a pronouncement concerning 
meaningful learning. A second weakness is the brevity of the experi- 
ments. Usually they are based on only a few days of training— a few 
months at the most. Optimal transfer— even normal amounts of trans- 
fer— cannot be expected under such conditions. In actual life, the 
development of skills and the building of personality take time; the 
process is continued for years. But whatever the demerits of the ex- 
perimental approach, it is still necessary to appeal to evidence, such 
as it is, from this source when examining the case for partial identity. 

If the theory of identical elements is to be defended at all, it 
must be defended in its literal meaning. It holds that mental life is 
composed of a certain number of well-defined habits, that these habits 
function independently excepting where some tangible, definable, 
testifiable “identical elements” connect them. This insistence upon 
“tangible, definable, testifiable” bonds is necessary in order to pre- 
vent proponents of the theory from slyly crawling out of the argu- 
ment. All too often champions of the theory change their meaning 
of “identity” or of “element” in the midst of their discourse. 

Parenthetically it may be added that, strictly speaking, the term 
“transfer” ought never be used by those who believe in partial iden- 
tity. Transfer suggests change of vehicle, whereas the hypothesis of 
identical elements should mean just what it says— identical vehicles. 
If one rides in the same vehicle throughout different territories one 
does not “transfer.” The excursion unit is always the same. If it is the 
selfsame habits that function in different contexts, there is, strictly 
speaking, no transfer at all. But this difficulty with terms may be 
waived. It is more important to expose the central weaknesses of the 
theory than to recover a term from misuse. 

16 P, T. Orata, Theory of Identical Elements, 1928, pp. 47-51- 




Thorndike likes ro think of an element as very specific indeed, 
e.g., “By identical elements are meant mental processes which have 
the same cell action in the brain as their physical correlate.” 17 Like- 
wise, “The general theory of identical elements— that one ability is 
improved by the exercise of another only when the neurones whose 
action the former, represents are actually altered in the course of the 
exercise of the latter— is sound, and is useful in guiding thought. 
However, so little is known about which neurones are concerned in 
any ability that this general theory does not carry us far.” 18 The 
image is one of identical neurone action. 

We are told by Thorndike that a person’s ability to estimate the 
length of 100 millimeter lines is essentially independent of his ability 
to estimate the length of jo millimeter lines. 19 Since in experimental 
investigations it is found that training in one of these abilities does not 
appreciably affect the other ability, they are regarded as containing 
no appreciable elements in common. One wonders whether the abili- 
ties involved in estimating lines of 100 mm. and 75 mm. are still 
separate; and how about the abilities to estimate lines of 100 mm. 
and 99.999 mm.? In another connection we are told that there are 
“as many memories as there are facts to be memorized”; each memory 
is regarded as an element in itself. But what is a “fact to be mem- 
orized”? Is it a syllable, a word, a sentence, a sequence of sentences, 
an entire poem verbatim, or the general sense of the poem? 

At the same time, extensive mental processes, such as “aims,” 
“moods,” “ideals,” “knowledge of procedure,” and such enormously 
complex functions as “esteem for truth wherever and however 
present” are alleged as elements. 20 If transfer can come about through 
such generalized components as these, the term element must lose its 
reference to “common cell action in the brain.” To propose that a 
generalized sentiment, such as “regard for the scientific method,” or 
“esteem for truth wherever and however present” has a rigid and 
specific location in the nervous system is to propose an absurdity. 

17 Educational Psychology , 1913, II, p. 359. 

ls lbid., II, p. 417. 

19 }. Phil., Psychol. & ScL Meth 1909, 6, pp. 239 f. 

20 Educational Psychology, 1903, p. 81, and Educational Psychology, 1913, H 
pp. 419 f. 



Such alleged elements are not at all specific; they are extremely 
general; in fact, they are the utmost in trait-psychology. 

At this point the conception of a truly elementary element is 
hopelessly lost, and for it is substituted the doctrine that transfer may 
be due to general attitudes, to concepts, or to traits, although this 
substitution is not recognized. It admits, without realizing the conse- 
quences, that whenever a generalized disposition is meaningful and 
relevant to two situations, it carries over from one to the other, exert- 
ing in both cases a determining influence upon behavior that makes 
the pattern of adjustment to one situation correspond in a meaning- 
ful way to the pattern of adjustment in the other. This is all that trait- 
psychology would maintain. 


The question now arises as to where identities occur in the 
sequence of events from stimulus to response. Conceivably the identi- 
ties might occur in the successive stimulus fields, or in neural proc- 
esses, or in ideational activity (presumably reducible to neural proc- 
esses), or in the final muscular contractions— or else in all these 
phases of the transfer situation. 

Most champions of the theory stress identities in the stimulus 
fields. To react to identities in the midst of diversities, they say, is 
the crux of transfer (cf. Figure 19, a and a?). But is an a actually 
identical with an /? Is one door identical with any other door? Do 
not these doors take their perceptual character rather from the whole 
situation in which they are embedded? A panther behind a cage is 
certainly a very different stimulus from a panther in front of the 
same cage. So it is with doors. The door to a neighbor’s house can 
never be identical with the door to one’s own house— neither ob- 
jectively nor subjectively. Even the most persistent facts of nature, 
such as the sun setting in the west, or the shape of the figure 8, are 
never precisely the same in all respects. And especially are they dis- 
similar in their effects upon the organism at different times, depend- 
ing on organic states of fatigue or desire, and on the contexts in which 
they occur. 

The most one could argue in favor of objectively “identical” ele- 
ments is that they sometimes seem to be equivalent for the purposes 
of the organism— functionally they can be substituted for one 


another. One friend may replace another as confidant or as partner 
at bridge. But in such cases it is not the objective identity that is 
responsible for the continuance of the same behavior, but rather the 
needs, attitudes, and traits of the person. It is these general disposi- 
tions that carry over from one situation to another in spite of the 
lack of identical elements in two stimulus situations. 

Even if there were ascertainable identities in two stimulus fields, 
they certainly could not be regarded as automatically provocative of 
transfer. According to the theory, identities in the stimulus fields 
should of themselves arouse the habits to which they are bound. Yet 
it took centuries for men to discover the “identical element” in the 
behavior of the falling apple and the tides. Usually it takes years for 
students of psychology to see the manifestly common features be- 
tween diverse schools of thought. One is seldom affected spontane- 
ously by so-called identities. The scientist must search for identities 
in the midst of diversity. Only by hard work, and skillful analysis 
can he find identities (or equivalences), by which he may transfer 
his knowledge, acquired from one object, to like objects. There is 
nothing compulsive about identities; they do not automatically cause 
transfer. Only under the guidance of general concepts and volitional 
dispositions are they effective. 

As for identities in the nervous system, they are even more elusive. 
Yet the belief that identical elements reside in identical cell action 
is basic to the whole theory. “In the same organism,” writes Thorn- 
dike, “the same neurone-action will always produce the same re- 
sult. . . 21 The picture he gives is one of definite neural grooves, 

capable of re-excitation in different response-patterns. Today such a 
picture is wholly untenable. Its fallaciousness from the point of view 
of cerebral physiology, and its complete inapplicability to the phe- 
nomenon of transfer, are disclosed by Lashley: 

The doctrine of isolated reflex conduction has been widely influen- 
tial in shaping current psychological theories. ... If learning is re- 
stricted to particular synapses, there can be no influence of training 
upon other activities than those actually practiced; any improvement 
in unpracticed functions must be the result of nervous connections 
which they have in common with the practiced activities. The rejec- 
tion of doctrines of formal discipline seems to have been based far 
more upon such reasoning than upon any convincing experimental 

21 E. L. Thorndike, Educational Psychology, I, 1913, p. 7. 



There is no evidence to support this belief in identity of nervous 
elements. On the contrary, it is doubtful if the same neurons or 
synapses are involved even in two similar reactions to the same stim- 
ulus. Our data seem to prove that the structural elements are relatively 
unimportant for integration and that the common elements must be 
some sort of dynamic patterns, determined by the relations or ratios 
among the parts of the system and not by the specific neurons activ- 
ated. If this be true, we cannot, on the basis of our present knowledge 
of the nervous system, set any limit to the kinds or amount of transfer 
possible or to the sort of relations which may be directly recognized . 22 

As for identities of muscular contraction, they are still less suit- 
able to serve as a basis for transfer. There are many ways of perform- 
ing the same act; these ways may all be equivalent to one another, 
but clearly the muscular contractions are not identical. Writing with 
a crayon held in the right hand, left hand, toes, or teeth shows trans- 
fer effects, but the muscles involved in these cases are entirely differ- 
ent. 23 Furthermore, the acquisition of a specific muscular skill never 
helps in solving new problems in which the skill would be useful, 
unless the subject sees the pertinency of the skill to the novel situa- 
tion (cf. p. 278). The employment of the same muscular movements 
is more likely to be a consequence of transfer, rather than a cause. 

It seems, then, impossible to discover any “identities” sharp 
enough, tangible enough, or identical enough (!) to justify this 
theory of transfer. The theory is entirely explicit; it says identities 
and it means identities, but identities seem always to elude discovery. 
Thorndike himself is somewhat disturbed by this elusiveness, and 
speaks of “identities beyond our cognizance,” but he believes that 
there is little trouble in reaching an “approximate decision” in those 
cases where training is of practical importance. This “approximate 
decision” is not so easily reached as Thorndike thinks. Furthermore, 

22 K. S. Lashley, Brain Mechanisms and Intelligence , University of Chicago Press, 
1929, pp. 172 f. 

23 Considerable evidence proving that motor consistency in the expression of 
personality cannot possibly be due to the employment of identical muscular con- 
tractions, or for that matter, to identical nerve tracts is contained in G. W, Allport 
and P. E. Vernon, Studies in Expressive Movement , 1933. A typical finding: “The 
length of the walking stride and the area of the subject’s normal writing correlate, 
r = -f.46. In such dissimilar fields of behavior it clearly cannot be a question of 
identical nerve processes. It seems that each subject is simply maintaining what for 
him is a suitable and congenial level of activity. Somewhere between the minimum 
and maximum possible extent, there is an extent of movement natural to the sub- 
ject, and this remains proportionately constant in very diverse situations whatever 
musculature is employed. The theory of identical elements is not equipped to handle 
a case of preserving a congenial proportionality in movement” (p. 156). 


an approximate decision is not enough. If the theory is to be defended 
at all, the identities must be specifiable and not everlastingly shifting 


Thorndike’s influential theory holds not only that transfer de- 
pends upon the presence of identical elements, but that the amount 
of transfer depends upon the nimber of identical elements present in 
two or more situations where transfer occurs. It is relatively simple 
to put this corollary to experimental test, although to determine the 
number of common factors involved it is again necessary to rely upon 
the “approximate decision,” endorsed by Thorndike. To reach such a 
decision probably alb would agree, for example, that the learning of 
two poems involves more processes in common than the learning of 
one poem and one set of nonsense syllables. However, in precisely 
this situation Sleight found no evidence of proportional transfer. A 
group trained in poetry showed as great an improvement in ability to 
memorize nonsense syllables as in ability to learn new poetry. 24 The 
results are typical of experiments on transfer. 

Even the original experiment of Thorndike and Woodworth 
showed an embarrassing absence of proportional transfer. 

Individuals practiced estimating the areas of rectangles from 10-100 
sq. cm. in size until a very marked improvement was attained. The 
improvement in accuracy for areas of the same size but of different 
shape due to this training was 44 per cent as great as that for areas of 
the same shape and size. For areas of the same shape, but from 140- 
300 sq. cm. in size, the improvement was 3 per cent as great. But for 
areas of different shape and different size (140-400 sq. cm.), the im- 
provement was 52 per cent as great. 25 

The greatest transfer was secured with areas that differed in both 
size and shape from practiced areas! Results so contrary to expecta- 
tion fail completely to justify the hypothesis of proportional transfer. 

Other evidence is plentiful. Coover, after excluding so far as pos- 
sible identical elements in two situations, still found transfer ranging 
from 25 per cent to 75 per cent; in other experiments he found trans- 
fer between dissimilar data greater than between similar data. Like- 
wise in the work of Cole, Coover and Angell, McKinney and many 

24 W. G* Sleight, Brit. j. Psychol., 1911, 4, 386-457. 

25 Cf. E. L. Thorndike, Educational Psychology , 1913, II, p. 397. 



others the proportional transfer required by the theory is lacking . 28 

Not only in sensory-motor learning is the amount of transfer 
quite independent of the proportion of identical elements in two 
stimulus fields, but the same lack of ratio is clearly seen in other forms 
of learning as well. Take trauma, for example, where a single episode 
influences all the recesses of a life, not merely those habits to which 
it has been previously bonded. A traumatic change of one small item 
in a man’s environment may make ail the difference in the world 
to him, perhaps changing happiness into despair. What happens in 
trauma is that a single experience saturates the whole life and forcibly 
compels a re-structuration of the entire personality. There is no trace 
whatever of proportional transfer. 


A classic experiment on transfer was that of E. Ebert and E. 
Meumann . 27 These investigators found such widespread improvement 
in the ability to memorize all sorts of material after training with 
nonsense material, that they felt compelled to postulate a Mitiibung 
(a sympathetic exercise or spread) in all memorial capacities follow- 
ing training in one special type of memorizing. G. E. Muller objected 
to this interpretation as extravagant, claiming by contrast that various 
identities (of Lemstoff, Lemmittel, and Lernweise) would account 
equally well for the results . 28 Ever since this controversy it has been 
customary for the defenders of partial identity to join Muller in 
claiming, whenever identities of content are “beyond cognizance,” 
that certain identities in procedure must be responsible for the results. 
Among these alleged identities of procedure are such complex mental 
states as “being aware that one has a problem,” “distrust of opinion,” 
“disposition to neglect irrelevant things,” or the regarding of certain 
courses of action as desirable, beautiful, or false . 29 

Now such extensions of the theory of partial identity are fatal; 

20 L. W. Cole, /. Educ. Res., 1928, 18, 32-39; J. E. Coover and F. Angell, Am. J. 
Psychol., 1907, 18, 328-340; F. McKinney, J. Exper. Psychol., 1933, 16, 854-864. 

L. W. Webb correctly observes that objective similarity of stimulus patterns 
may not mean similarity of the neural activities involved, but adds, “To my mind 
herein lies the weakness of Thorndike’s theory. Its validity can never be adequately 
tested. Any general agreement as to the degree of neural identity between any two 
complex problems is impossible.” ( Psychol . Monog., 1917, No. 104, p. 57.) 

2T Arch. f. d. ges. Psychol., 1905, 4, 1-232. 

28 Zsch. f. Psychol., 1905, 39, iii-iij. 

29 E. L. Thorndike. Educational Psychology, 1913, II, pp. 418 f. 



they mean nothing less than its complete surrender. Take the “identi- 
cal procedure,” so often mentioned, of “getting to the heart of the 
problem directly.” Such a complex mental attitude varies enormously, 
depending on the stuff of the problem. The irrelevancies that must 
be excluded in approaching the heart of different problems are not 
identical irrelevancies. “Habituation to discomfort and fatigue,” 
“attentiveness,” and “the disposition to maintain the scientific atti- 
tude,” are never twice exactly the same. To maintain a scientific atti- 
tude, for example, requires many different associations, movements, 
and mental operations. The only common factor is a thoroughly 
generalized attitude or interest, versatile in expression, employing now 
this neural mechanism and now that, characterized by more flexibility 
than the theory of identical elements can admit with consistency is its 
own position. 

The inclusion of general attitudes and volitional dispositions 
under the list of “identities” is a necessary but futile subterfuge to 
save the theory. The explicit thesis is that 

Training the mind means the development of thousands of particu- 
lar independent capacities, the formation of countless particular ca- 
pacities, for the working of any mental capacity depends upon the 
concrete data with which it works . 30 

This is the thesis and its proponents must stick to it. The scientific 
attitude, or a disposition to search for what is beautiful, or a general 
regard for truth, admitted by Thorndike as identities of procedure, 
are obviously not “particular independent capacities” specific to the 
concrete data to which they are applied! 

Procedure in learning is indeed a crucial factor in transfer, but 
procedure depends primarily on attitude, interest, aim, concept, ideal, 
or an understanding of principles. All of these forms of mental organ- 
ization result in transfer. But none of them is constituted of invariant 
habits or specific bonds. 

Thorndike once found that certain pupils taught to square x + y 
suffered a great deal of interference when asked to square bi + b 2 . 
He concluded that squaring x + y is an ability specific and inde- 
pendent of squaring bi + b 2 . 31 Is it not perhaps the teaching that was 
at fault? A child often fails to see the principle of a computation, but 
the instruction is usually to blame. It is even possible to teach a young 

30 E. L. Thorndike, Principles of Teaching , 1906, p. 24.8. 

31 /. Exper . PsychoL , 1922, 5, 33-38. 


child in such a way that his ability to count three pennies will not 
extend to the task of counting three apples, but such mischievous 
pedagogy proves only that under unfavorable conditions generaliza- 
tion will not occur. 

The importance of forming a general concept in order to secure 
transfer, and the worthlessness of the blind recurrence of identical 
elements, is demonstrated in the well-known experiment of H. A. 

A certain mechanical puzzle was so arranged that it could be pre- 
sented in various forms. The manipulations for these various forms 
could all be comprised under a single formula. This general formula 
could be deduced from any one of these special forms. A number of 
subjects were tried with this puzzle. As soon as skill was acquired in 
dealing with one form of the puzzle it was changed to another form. 
The subjects who developed the general formula during the solution 
of the first form were able to use the specialized habits built up in the 
first form in the second. Those who formed merely the special habits 
without developing the principle attempted to carry over the habits 
without modification and were greatly embarrassed by the change . 32 

In another experiment, a group of students was taught geometry 
in the prevailing routine fashion, and another group by an improved 
technique devised to emphasize logical principles. The groups were 
equal in intelligence. After some months, the experimental group 
showed more persistence, developed more methods of solving new 
problems, secured higher examination grades, and even showed im- 
provement on non-geometrical reasoning tests . 33 This experiment 
shows that as ordinarily taught geometry does not carry over to 
reasoning in general, but likewise that it may be so taught as to 
achieve this end. Many other investigations lead to essentially the 
same conclusions . 34 One study showed that training in neatness may 
or may not transfer to activities other than that in which neat habits 
were first established. Everything depends on the way training is 
given . 33 Transfer is not automatic, but depends upon insight into 
general principles. Good teachers know this is true . 36 

32 “The Psychology of Efficiency,” Arch, of Psychol., No. 15, 1910, 18 f. 

33 E. P. Johnson, Mathematics Teacher , 1924, 17, 191-201. 

34 E-g H. Woodrow, /. Educ. Psychol 1927, 18, 159-172; G. P. Meredith, 
Formn of Educ., 1927, 5, 37-45; W. W. Coxe, /. Educ. Res 1923, 7, 244-247. 

35 W. C. Ruediger, Educ. Rev., 1908, 36, 364-371. 

36 “The conclusion is inevitable that when an individual is trained in mere rou- 
tine fashion or drill, he gets fixed and mechanical habits which do not transfer. On 
the other hand, when he is trained consciously to organize his knowledge or pro- 




The theory diametrically opposed to the doctrine of partial iden- 
tity is known, somewhat broadly, as the theory of generalization. Evi- 
dence for it is plentiful enough, but its detailed formulation gives 
more difficulty. 

The experiment of C. H. Judd that led him and others to deny 
the theory of identical elements, and to offer a theory of generaliza- 
tion in its place, is well known. Two groups of boys were directed 
to throw darts at a target 12 inches under water. After one group 
had received instruction in the principles of refraction, the target was 
then placed four inches under water, and all the boys tried again. 
On this retest the uninstructed group had larger and more persistent 
errors, while the boys who had learned the theory adjusted very rap- 
idly to the changed conditions . 87 Later, Alpert, working with pre- 
school children, found that comprehension (insight) of a solution im- 
proved the subsequent solution of similar tasks, whereas trial and error 
solutions did not . 38 Ruger (whose experiments were cited above) 
concluded that “In general, the value of specific habits under a change 
of conditions depended directly on the presence of a general idea 
which would serve for their control .” 88 Pear, with tasks involving 
low-grade motor skill in which conscious generalization- was made 
impossible by the conditions, found little transfer. He concludes that 
there are two meanings of transfer, (a) that resulting from exercise 
of any particular function in two contexts (Thorndike’s theory) ; and 
(b) transfer resulting from the extension of attitudes, sentiments, 
ideals, or knowledge of methods, where the particular function 
trained is a vehicle of these mental powers. Lie writes, “It now seems 
certain that (a) is rare, and that (b) definitely can occur .” 40 

There is no need to multiply evidence. Time and again it appears 
that identical elements in themselves have no power to effect transfer. 
Only when a general principle is understood as applicable to two or 
more fields does the training in one carry over to the others. Fre- 

cedure in such a way that general principles are formulated, the result is not a me- 
chanical habit but generalization, or an adaptive and flexible form of behavior which 
by virtue of its flexibility transfers.” P. T. Grata, The Theory of Identical Elements. 
1928, p. 99. 

37 C. H. Judd, Educ. Rev., 1908, 36, 28-42. 

38 A. Alpert, T. C. Contrib. to Educ., No. 323, 1928. 

39 H. A. Ruger, op. cit., p. 19. 

*°T. H. Pear, Nature , 1928, 122, 611-614. 



quently this transfer occurs even when the subject is ignorant of 
specifically identical features in the various situations. He need have 
only a valid conceptual principle in mind. When introspections are 
taken it often turns out that the subject discovers identical features 
between his problems only after the transfer has taken place, and 
then merely because he is looking around for some means of rational- 
izing his first valid generalization. 

One form of generalization aiding transfer, but utterly contradict- 
ing the hypothesis of identical elements, is the carrying over of an 
abstract relationship from one situation to another. For instance, in 
Kohler’s work, many times verified, an animal trained to respond to 
the blighter of two grays by receiving food when the brighter is 
selected, will, when confronted with a new pair of grays, forsake the 
same gray (the specific element) with which food has been pre- 
viously associated, in favor of one that is still brighter . 41 It is not the 
identical element that provokes the selection; the idetitical element 
with which food has been associated is actually abandoned. Monkeys, 
too, will choose the brighter, the squarer, the louder, the heavier, or 
the larger of two objects in their attempt to keep up with the experi- 
menter’s scheme for feeding them. It is impossible, of course, to argue 
that the relation between two stimuli is itself an element identical 
with the relation between two different stimuli, because in every such 
case wholly different sensory processes are involved . 42 

1 : { 

it || 

;:ji if 


To prove that transfer occurs, not through the re-activation of 
specific bonds, but through the perception of relations, through in- 
sight, and through the formation of general concepts, is important, 
but not altogether definitive. There are limitations to any individual’s 
powers of generalization, and there are conditions and principles gov- 
erning its occurrence, all of which need to be known, before the 

41 W. Kohler, Abhandlungen der Preuss. Akad. d. Wissenscbaft, 1918, Phys- 
Math., Kl., 2; also HL Kliiver, Behavior Mechanisms in Monkeys, 1933. 

42 The case for partial identity looks even more hopeless in the light of Lash- 
ley’s report on the transference of the ability to discriminate brightness from one eye 
to the other which had not been used during training. In this experiment the same 
sensory cells were certainly not stimulated in the formation of the habit. More re- 
markable still is the fact that this transfer took place in animals without a visual 
cortex, showing further that “identical cell action” is hardly the basis for the trans- 
fer. (The Foundations of Experimental Psychology, ed. by C. C Murchison, 1929, 
P* 543 *) 


nature of transfer, or of the structural units of personality, can be 

Much help comes from the experimentalist’s conception of equiv - 
alence of stimuli . This conception grew out of the discovery* that 
changes in the environmental situation, sometimes even pronounced 

B B' 

Figure 21 

Equivalent and Non-equivalent Stimuli 

Monkeys trained to respond positively to the square and to avoid the circle 
will transfer these responses to a wide range of equivalent relationships. These 
designs are reproduced through the courtesy of the experimenter, H. Kliiver. 

changes, do not essentially alter the nature of the response. 48 For ex- 
ample, night monkeys (Aotes), trained to respond favorably (in seek- 
ing food) to shape A (Figure 21) and to avoid shape B, will, when 
confronted with A' and B', select the former and avoid the latter. 
For them A and A', B and B' are equivalent. 

To explain such remarkable transfer effects, the theory of identi- 
cal elements helps not at all. There are no conceivable elements or 

43 H. Kliiver, op. cit p. 4. The same author has given an historical account of 
the method of equivalent stimuli, and has pointed out its value in the investigation 
of personality, Char. & Pers. } 193 6, 5, 91-112. 

28 o 


identities involved unless they be curvilinearity and angularity , but 
each of these is a relationship rather than an element. Further, the 
monkey responds to neither as separate from the other but only to a 
highly generalized cue, viz., greater curvilinearity . That the transfer 
is due to such a magnified degree of generalization is proved through 
the use of a large number of pairs of stimulus figures, having abso- 
lutely nothing in common save this greater and lesser degree of 
curvilinearity. The phenomenon which Kliiver calls equivalence, 
therefore, is one of generalization. 

The same principle can be applied to the dispositional units of 
personality. Suppose, for instance, one is studying the personality of 
a “super-patriot.” One quickly discovers that to him many different 
items of experience are equivalent in the response they provoke: a 
red flag, a volume by Marx, a teacher’s union, the peace efforts of a 
Yale professor, or the formation of a neighborhood co-operative. 
One and all they arouse the communist phobia. The equivalence of 
these items can be explained only by assuming that there is an under- 
lying disposition of some sort, whose threshold of arousal is low and 
capable of being crossed by this whole diverse range of stimuli. How 
can one explain the fact that such diverse stimuli are equivalent (i.e., 
not responded to with discrimination) unless one postulates a general- 
ized personal disposition that renders them equivalent? This inference 
becomes all the more necessary when one discovers that diverse re- 
sponses to these stimuli are likewise equivalent. The super-patriot in 
question may, on occasion, write a letter of protest to the local news- 
paper; on another, merely grow red with anger and bluster; or, again, 
he may join the vigilantes. What is manifestly constant is the attitude ; 
it is because of the attitude that many stimuli and many responses 
are equivalent. Neither in the stimuli arousing the attitude, nor in the 
responses issuing from ir, nor in the “cell-action of the cortex” is it 
possible to trace identical elements. The memmgfulness of the dis- 
position, its significance in the life-pattern of the super-patriot, is the 
one and only stable feature. 

The degree of generalization of such a disposition (which may 
here be called either a trait or an attitude) is measurable by the range 
of equivalence. Some people with an anti-communist attitude do not 
treat so wide a range of experiences as equivalent (with so little dis- 
crimination). Some likewise have a much narrower behavioral range 
within which responses may be regarded as equivalent. The degree 



of generalization, which is different in different persons, varies in- 
versely with the degree to which stimuli and responses are discrim- 
inated. A finely discriminated disposition, one that is aroused only by 
some narrow and special class of stimuli, leading to a special kind of 
verbal or motor behavior, is likewise a unitary structure in person- 
ality, but far more specific than the widely generalized attitude just 
described . 44 

The welter of stimuli to which the individual is exposed fall for 
him, as it were, into constellations, every member of which is effec- 
tive in producing some response. Correspondingly, the responses he 
makes, though almost infinitely varied, are not as diverse as they 
appear at first sight, for many of them are also equivalent in their 
personal significance. Thus, for a man with a disposition to be polite, 
innumerable environmental occasions are equivalent in their power 
to arouse this particular determining trait, and at the same time the 
polite gentleman finds innumerable ways of expressing his dominant 
trait (equivalent responses). 

The application of the doctrine of equivalence to the^ problem of 
transfer is easily made. Transfer takes place whenever, and in what- 
ever degree, a new stimulus field for a given individual is equivalent 
to another more familiar field. A child learning nearness will carry 
over this orderly behavior from the training situation to a new situa- 
tion to the extent that the two situations are for him equivalent. If to 
him the fields are non-equivalent, there will be no transfer. And if in 
some way, the fields are conflicting or antithetical, negative transfer, 
or interference, will occur. 

Related to this modern concept of equivalence is the ancient con- 
cept of similarity. For centuries “association by similarity” was re- 
garded as a primary law of mental organization. In recent times, 
under the attack of specificity, partial identity, and related elemen- 
taristic crusades, this law lost its standing and was reduced to identi- 
ties and contiguities. If, as we have argued, this reduction is impossi- 
ble, then the concept of similarity again raises its bruised head. 

To enter here into a full examination of this difficult problem 
would be impossible. Its reformulation with the aid of the many 
experiments on the equivalence and constancy of stimuli, and the 

M Though both specific and general attitudes must be recognized as true dispo- 
sitions, the latter type is ordinarily superior in duration and in motivational power. 
Cf. H, Cantril, “General and Specific Attitudes,” Psychol. Monog No, 192, *93 2 - 



equivalence and consistency of responses, is now gradually taking 
place. It is still premature, however, to regard the concept as fully 
reinstated to a central position in the explanation of transfer, or in 
the analysis of personality. But its status is improving. 

Formulations in terms of equivalence are essentially behavioristic 
and operational. The investigator determines the range within which 
stimuli are equated, and the range within which responses are con- 
sistent. But the account in these operational terms is not complete 
without a corresponding formulation in phenomenological terms; for 
the experience of similarity, subjectively considered, is remarkably 
subtle and its role in mental organization cannot as yet be stated 
entirely in terms of operational equivalence. 

Indeed the capacity of the human mind for sensing resemblances 
is astounding. Whenever two stimulus fields (however novel and 
different they may be) are for any perceiving mind, for its own 
purposes, classed together, a sense of similarity results. It is a highly 
individualistic matter, with the oddest of metaphors sometimes result- 
ing. Someone’s voice may be sensed as similar to blotting paper, some- 
one’s eyes as similar to twin ash cans just emptied, someone else’s 
moral fiber as like a sponge. The capacity of mind for forming such 
analogies is almost limitless. Small wonder that William James was 
prompted to speak of the “lawless revelry of similarity.” 

For all its individuality of form, and for all its unpredictability, 
the sense of similarity is the subjective condition par excellence of 
transfer. If a student sees similarity between the Greek grammar at 
which he drudges and his own love of logic, mutual transfer results; 
if not, there is no such transfer. If for him there is some meaningful 
link between his instructor’s criticism of his errors in English com- 
position and his sloppy conduct outside the classroom, again transfer 
occurs (cf. p. 210). 

To some students the study of English, psychology, or even of 
the much despised Latin, may have many valuable resemblances to 
life. For others these subjects are barren, not because they fail to 
possess elements that specifically overlap with life outside the class- 
room, but because the persons can see no similarity. For this group 
more “practical” subjects, such as business English or advertising, 
may kindle the undying flame. 

Similarity is, then, seen to be a highly individual matter. Its opera- 
tion is difficult to predict, and educationally it cannot be standard- 


ized. 45 To be sure, a teacher often points out resemblances between 
the experiences of his students inside and outside the classroom, but 
the results are unpredictable. Some remain impervious to the effort; 
others who have what James called “the electric aptitude for analo- 
gies,” may perceive the similarities that the teacher points out and 
many more besides. 

But is this revelry of similarity really lawless ? It is conditioned 
by temperament and previous learning; it is certainly lawful enough 
if regarded from the point of view of the individual’s life-history. 
Instead of lawless it would be more accurate to say that similarity is 
personal. And just because it is personal, the transfer value of any 
item in the school curriculum cannot be standardized or predicted. 
Neither for school children nor for adults are similarities uniform. 
Only in the sense that the transfer value of similarity evades standard- 
ization for the group is it lawless; in the sense that it always harmo- 
nizes with the individual life-history it is lawful. 

Since similarity is essentially the subjective counterpart of equiva- 
lence, it plays a corresponding role in the structural arrangement of 
personality. To an artistic person, a sunset, a landscape, a sonnet, a 
threshing machine, a derelict in the sands, may all be similar: in that 
to him they all mean beauty. The wider the range of such conscious 
similarities, the more generalized is his esthetic interest. 

It is not necessary to demonstrate again that such similarities 
cannot possibly be reduced to partial identity. Resemblances are 
never identities, not even in certain respects. If I say that you remind 
me of some friend, and you ask me why and wherefore, I may narrow 
down the resemblance to similar build or similar resonance of voice. 
But even in these respects you are not identical with my friend; your 
build or your voice is only similar to his. The search for the partial 
identity involves an infinite regress. Even should we come by analysis 
to some minute feature that seems virtually interchangeable between 
you and my friend, we then are forced to admit that this residual 
feature does not represent the original experience which was of a 
resemblance between him and you as 'whole human beings. 

Every experience of similarity implies a paradox. There is both 
sameness and novelty in the experience. Two stimulus fields are com- 
pared, and though for certain purposes they may be subjectively 
equivalent, still they are known to be disparate and non-identical. 

45 And under certain special conditions of conflict and confusion, similarity may 
lead to negative transfer. 


A sense of two-ness, not of identical one-ness, is always involved. 
The sense of novelty and of separateness in experiences of similarity 
enables us to escape from literal and stereotyped response. Even while 
we are reducing the limitless and confusing events of life to a man- 
ageable number— seeing analogies here and likenesses there— we find 
plenty of room for originality and discrimination in our adjustments. 

The principles of equivalence and similarity (objective and sub- 
jective counterparts of the same mental process) are a great improve- 
ment over the theory of identical elements. The latter is a rigid, 
mechanical principle allowing only invariable “reactive” response; 
the former allow both for stability and for flexibility in conduct. As 
the individual develops he learns to economize his responses; a wide 
range of environmental demands take on, for him, the same signifi- 
cance (becoming equivalent and similar). Concepts are formed, like- 
wise attitudes, and traits: these are for him stable modi vivendi, 
serving him as dependable principles of survival in a confusing world. 
Yet these concepts, attitudes and traits are not rigidly bound to the 
stimulus according to the principle of identical elements. If they were, 
a crude stereotyping of behavior would ensue, and personality would 
be far more machine-like than it is. 

The reader can now see why this discussion has been a necessary 
prelude to a new and more adequate theory of the structure of 
personality. The true units of personality must be serviceable for 
survival. They must allow for both stability and flexibility in be- 
havior. Even while the individual, through training, learns to behave 
to wide ranges of stimuli in essentially the same way (perceiving 
them as similar) , at the same time he must constantly meet and master 
new demands. Hence, the units of personality, well structured though 
they are, cannot be fixed and changeless. 


The solution of the problem of the organization of personality 
will depend to a large degree upon the solution found for the prob- 
lem of transfer. Both face the difficult question as to how the indi- 
vidual meets the succeeding events of his life through employing 
his relevant previous experience. 

During the past generation the prevalent view of transfer has 
been in terms of hypothetical identical elements; and this doctrine, 
at the same time, has been widely accepted as an adequate theory 



of the structure of personality. Everything mental is said to be ruled 
by habit, and habits are conceived as insulated stimulus-response 
bonds, affecting one another and combining into patterns only to the 
extent to which they have some still more specific elements in 

Such a conception must be rejected. Either, on the one hand, the 
search for elements becomes a matter of infinite regress, more and 
more minute (and more and more elusive) entities being sought; or 
else, on the other hand, the literal intent of the theory is sacrificed 
by positing highly generalized elements, such as “desire for truth,” 
or “an attitude of going to the heart of any problem.” In the latter 
case, no specific identities can be found. In spite of the explicit re- 
quirement of the theory, it is never able to specify the precise nature 
or location of any identity. 

On the experimental side there is no evidence for proportionality 
of transfer, that is to say, for the contention that transfer occurs in 
proportion to the number of approximate identities between two 
fields of stimulation. The falseness of this contention is especially 
apparent in all cases where trauma or emotional learning is involved. 
In these instances, transfer passes all bounds of expectation. In such 
cases identities cannot be involved, for the whole personal life is 
saturated with the effects. All in all, experimental data, the discoveries 
of modem neurophysiology, the canons of theoretical psychology, 
and simple common sense, unite in rejecting this view of mental 

Evidence favors a theory of the opposite order, one in which 
integration and generalization play the leading part. Here transfer 
effects depend chiefly upon the equivalence of meaning to the indi- 
vidual of the fields that confront him. If they are similar, transfer 
takes place. Equivalence and similarity are not uniform for all; they 
are a personal matter, and hence it is impossible to predict for people 
en masse the transfer value of a single experience, or to arrange a 
program of school studies that will secure uniform transfer effects 
for all children. 

To explain equivalence and similarity it is necessary to assume 
underlying focal dispositions within the latent mental organization 
of each individual. These dispositions must be of such an order as to 
account both for the stability and versatility of response. Upon the 
ground thus cleared, a constructive and detailed theory of the struc- 
ture of personality may now be built. 



We have the right and the obligation to develop a concept 
of trait as a definitive doctrine; for in all the activity of the 
Person, there is besides a variable portion, likewise a constant 
purposive portion, and this latter we isolate in the concept of 

—William Stem 

Scarcely anyone has ever thought of questioning the existence of 
traits as the fundamental dispositions of personality. In common 
speech everyone presupposes traits when he characterizes himself or 
his acquaintances. This man, one says, is gruff and reticent, but a 
hard-worker; that man is fastidious, talkative, and penurious. Nor- 
mally the psychologist too talks in these terms. But as soon as he 
enters his laboratory or class-room he is likely to leave common sense 
behind him and to embrace one or another of the scientific theories 
discussed in the last two chapters. Rightly he thinks of common sense 
as a faulty guide. Yet in the matter of human traits common sense 
is remarkably well-seasoned with experience, and scarcely deserves 
the complete rebuff it receives. 

For one thing the path of common sense would never have led 
into the thickets of specificity nor run the aimless course of the doc- 
trine of identical elements. Such non-focal theories of organization 
are entirely foreign to it. On the other hand, common sense would 
not go to the opposite extreme and regard personality as a perfect 
unity, faultless in integration. It is between the levels of specific 
habit and complete unity that common sense locates the natural foci 
of personality. It is there that psychologists should search . 1 

1 And physiologists likewise. Up to the present time there has been little work 
beyond rough speculation concerning the dynamics of “traces,” of “cortical stresses,” 
and “neurograms,” all defined as multi-focal dispositions ready-organized to perform 
in some integrative fashion either at high tension or at low. (Cf. E. B. Holt, Animal 
Drive and the Learning Process, 1931; see especially pp. 167, 161, 263 for suggestive 
statements regarding this in-between level of analysis.) But in general so little is 



A certain metaphysical question is likely to arise at the outset 
of any discussion of traits, and if allowed to do so it bedevils the 
whole problem. The sooner it can be disposed of, the better. The 
question, in brief, is this: Are traits bona fide , veridical dispositions, 
or are they nothing more than nominal fictions, mere words, obscur- 
ing rather than clarifying the structure of personality? Some writers 
maintain the first position, some the second, and a few occupy a 
somewhat uncertain middle ground. The following quotations are 
typical of these three positions. 

My own view is that traits are only convenient names given to types 
or qualities of behavior which have elements in common. They are 
not psychological entities but rather categories for the classification 
of habits. 2 (biosocial) 

A trait is a constant directing psychic force ( Richtkraft ) which de- 
termines the active and reactive behavior of the individual. 3 (bio- 

An individual is said to possess, or to be characterized by, a certain 
personality trait when he exhibits a generalized and consistent form, 
mode, or type of reactivity (behavior), and differs (deviates) suffi- 
ciently from other members of his social environment, both in the 
frequency and intensity of this behavior, for his atypicality to be 
noticed by relatively normal and impartial observers, themselves mem- 
bers of this same environment. . . . The definition tries to express the 
notion that the trait is a relation between the individual and his ob- 
servers. 4 (mixed) 

The first quotation is compatible with the hypothesis of identical 
elements. If habits, connected by their common bonds, are the true 
elements of personality, then it follows, as the author asserts, that 
trait-names are merely designations for categories of habits. He con- 
siders traits not so much as residing in the individual himself, as 
forms of perception for an observer to use. 5 

known concerning the divisions and gradients of the neural process that a neuro- 
physiological account of human traits is as yet scarcely possible. In time the story 
developed here from a psychological point of view will no doubt find its physiolog- 
ical counterpart. 

2 M. May, J. Soc. Psychol. , 1932, 3, p. 133. 

3 R. Baumgarten, Die Charaktereigenschaften , 1933, p. 15; also Brit. /. Psychol 
1936, 26, p. 290. 

4 P. E. Vernon, Psychol. Rev. y 1933, 40, 542 f. 

s The same author’s suggestion that the total personality likewise be defined in 
biosocial terms, as “a man’s social stimulus-value,” was rejected in Chapter II. 


Of course, traits do have stimulus value, but this does not mean 
that they are only “categories for the classification of habits.” The 
fact that people can perceive, appreciate, name, and react to, the 
conduct of others, is not a legitimate reason why we must conclude 
that these others are devoid of traits. The perception or inference of 
traits by outside observers is a problem by itself, requiring later dis- 
cussion (Chapters XVIII and XIX); but it has no essential bearing 
upon the question of the existence of traits. 

The biosocial view of personality is likewise artfully expressed by 
Pirandello who never tires of pointing out that a man’s personality 
seems to vary with the expectations and prejudices of his associates. 
To one he may be a saint, to another a sinner, a hated enemy or a 
trusted ally. Whoever chooses to view a character in a particular 
fashion is, in a sense, right (if he thinks he is). But wherein does the 
multiplicity lie? Surely not within the person himself. The variability 
lies rather in the fact that into their judgments of him many people 
project their own desires and biases. What is most noteworthy in 
research on personality is that different observers should agree as well 
as they do in judging any one person. This fact alone proves that 
there must be something really there, something objective in the 
nature of the individual himself that compels observers, in spite of 
their own prejudices, to view him in essentially the same way. 

Another version of the biosocial view is “fictionism.” One of its 
first representatives was Jeremy Bentham who was ever on his guard 
against substituting fictitious entities for real ones. Traits of person- 
ality he considered in the former category: 

Now disposition is a kind of fictitious entity, feigned for the con- 
venience of discourse, in order to express what there is supposed to 
be permanent in a man’s frame of mind. 6 

Bentham led a valiant fight against hypostatization through the reck- 
less use of names, and no one can deny that obscurantism may result 
from their careless use. But this danger is not in itself sufficient 
grounds for a sweeping denial of the existence of traits. It indicates 
only the need for discrimination. As a matter of fact, on the page 
following the skeptical quotation, Bentham himself lays down the 
conditions under which the existence of dispositions may be securely 

6 Jeremy Bentham, Principles of Morals and Legislation, 1879 ed., chap, ix, 
p. 131. 


A somewhat pragmatic compromise with fictionism is proposed 
by Thurstone who accompanies his doubts concerning the existence 
of traits with the admission that human beings certainly behave “as 
if” they had traits. 7 There is no real objection to this “as if” epis- 
temology provided it is applied equally to all the data of psychology. 
It becomes objectionable only if it is used as a faint-hearted apology 
for traits as if for some reason they are to be regarded as less genuine 
than other forms of mental structure. 

The second quotation on p. 287 represents a typical biophysical 
conception. A trait, it says, has more than nominal existence; it is 
independent of the observer, it is really there. Although it makes the 
unequivocal assumption that there are traits, or some corresponding 
neuropsychic dispositions, it leaves the problem of the criteria of 
traits and their discovery in particular cases to the critical procedures 
of psychology. In other words, this view does not hold that every 
trait-name necessarily implies a trait; but rather that behind all con- 
fusion of terms, behind the disagreement of judges, and apart from 
errors and failures of empirical observation, there are none the less 
bona fide mental structures in each personality that account for the 
consistency of its behavior. This view, the most worthy of endorse- 
ment, will be developed at greater length in succeeding pages. 

The third quotation starts boldly enough with the recognition 
that there are such things as “generalized and consistent” forms of 
reactivity, but ends in the irrelevant sphere of social perception and 
judgment. It is a mixture of the biophysical and biosocial points of 
view. The first part of this definition is wholly satisfactory, but the 
biosocial additions are not. Why must a man’s atypicality be notice- 
able in order that traits be attributed to him? Doesn’t a person who 
is in no way outstanding or atypical possess traits as authentic as do 
conspicuous deviates? Distinctiveness is not the criterion of a trait. 
The author of this quotation likewise errs in assuming that a trait 
must be perceived by others, that it must represent some kind of 
relationship between two people. Did Robinson Crusoe lack traits 
before the advent of Friday? Will the last man to remain alive on 
earth abruptly lose his traits when his companions die? 

7 L. L. Thurstone, The Reliability and Validity of Tests , 1932, p. 101. 




Psychology has never been able to do without some conception 
of determining tendency (implying a readiness for response) . With- 
out such a conception it could never pretend to account for the 
manifest stability and consistency of behavior and experience. Deter- 
mining tendencies, of course, are not matters of direct observation, 
but only of inference. The inference, however, is altogether com- 
pulsory, for not only do the observed data lead inevitably in that 
direction, but without it psychology could not advance beyond the 
stage of recording unintelligible discrete acts and separate states of 
consciousness. To accept traits requires no radical revision of the 
psychologist’s creed, for traits are biophysical in the same sense that 
determining tendencies, attitudes, or other dynamic influences have 
always been considered by psychology as biophysical. 

The phrase “determining tendency” has both a narrow and a 
broad connotation. In its narrower sense it refers specifically to a 
mental set that facilitates the solution of a special problem or the 
execution of a certain act. In its broader sense, it is any directive 
tendency or condition of readiness for response. The doctrine of 
traits may be ordered to this broader conception. All traits are di- 
rective tendencies, but conversely all directive tendencies are not 
traits. Some directive tendencies are far too narrow and specific in 
their reference, and too fleeting in time to satisfy the criteria of a 
trait. To be sure even a transient mental set may have some de- 
pendence upon personal traits, for traits often underlie the determin- 
ing tendency of the moment. But in themselves traits are more gen- 
eralized and more enduring, having less to do with fleeting mental 
sets than with lasting mental structures such as interests, tastes, com- 
plexes, sentiments, ideals, and the like. 

There are two familiar classes of determining tendencies— habit 
and attitude— with which the concept of trait must be compared in 

Trait and Habit. Ordinarily the term habit connotes an invariable 
and inflexible type of response following the recurrence of a definite 
stimulus situation with which it is, by experience and practice, tied. 
The two previous chapters have explained at some length why this 
type of unit cannot be accepted as the sum and substance of mental 
organization. The units of personality must be more versatile both in 


respect to the situations that arouse them, and in respect to the re- 
sponses they provoke. 

In recent years some writers have sought to escape from the 
picture (bequeathed by William James) of the specific habit as a 
narrow task-master, and to put in its place a modified view, the doc- 
trine of the generalized habit , or of a disposition that is not invariable 
in its mode of arousal or in its forms of expression. 8 This conception 
is sponsored by Dewey. While still employing the standard term, 
habit , he portrays a far more flexible disposition than was character- 
istic of usage in the generation after James. 

Repetition is in no sense the essence of habit. Tendency to repeat 
acts is an incident of many habits, but not of all. A man with a habit 
of giving way to anger may show his habit by a murderous attack 
upon someone who has offended. His act is nonetheless due to habit 
because it occurs once only in his life. The essence of habit is acquired 
predisposition to ways or modes of response, not to particular acts, 
except as, under special conditions, these express a way of behaving. 
Habit means special sensitiveness or accessibility to certain classes of 
stimuli, standing predilections or aversions, rather than the bare re- 
currence of specific acts . 9 

So far as it goes, Dewey’s formulation of the generalized habit 
is in every particular equivalent to that of traits. Dewey does not 
deny the existence of some independent and specific habits, but these, 
he sees, are generally to be ordered under more inclusive dispositions. 
So much organization and coherence in mental life leaves no choice 
but to recognize great systems of inter-dependent habits, comprising 
generalized dispositions that grow in flexibility as they develop. 

A young child may be regarded as forming a specific habit when 
he learns (with difficulty) to brush his teeth night and morning. For 
some years this habit may stand alone, aroused only by appropriate 
commands or by the appropriate environmental situation. With the 
passing of years, however, brushing teeth becomes not only auto- 
matic (as is the way of habits) but likewise firmly woven into a 
much wider system of habits, viz., a trait of personal clecmliness. (If 
a more behavioral designation of the trait is desired, one can speak 
quite accurately though less conveniently, of a generalized tendency 
to remove all manner of dirt from one’s person.) The adult is un- 

8 Cf. S. S. Colvin, The Learning Process, 1921, pp. 49 f. 

»J. Dewey, Human Nature and Conduct, 1922, p. 42. 



comfortable if he omits brushing the teeth from his daily schedule, 
not only because a single habit is frustrated, but because the omission 
violates a general demand for cleanliness. 

This example implies, quite correctly, that a trait arises, in part 
at least, through the integration of numerous specific habits having 
in common not identical elements, but the same adaptive significance 
to the person. In addition to this gradual integration and conceptual- 
ization, the influence of personal temperament on the formation of 
traits must not be overlooked. As pointed out in Chapter IV, some 
styles of generalization are far easier for certain people to learn than 
others, depending upon innate influences of temperament, intelli- 
gence, and constitution. But however acquired, a trait is always a 
fusion of habits and endowment rather than a colligation or chain 
of habits alone. 

By tracing the hypothetical history of another generalized habit, 
sociability, we can see the process more clearly. A young child find- 
ing that his mother is nearly always present to satisfy his wants, de- 
velops for her an early affective attachment (conditioning). But later 
other social contacts likewise prove to be conducive to this child’s 
happy and successful adjustment: playmates, for example, or family 
gatherings, or crowds at the circus. Unless markedly timorous in 
temperament, or fearful and shy because of experiences of punish- 
ment or public ridicule, the child gradually comes to seek people, 
rather than to avoid them. A trait (not an instinct) of gregariousness 
develops. The child grows eager for social intercourse; he enjoys 
being with people. When isolated from them for some time, he 
misses them and becomes restless. The older he grows the more ways 
he finds of expressing this gregarious interest. He seeks to ally him- 
self with groups of people at the lodge, at the theater, at church; he 
makes friends and keeps in touch with them, often entertains them, 
and corresponds with them. These separate activities are not habits. 
They are varied (but equivalent) aspects of a trait of sociability. On 
occasion this trait may become dynamic almost to the point of com- 
pulsiveness, leading to such excess of sociability that the person is 
morbid or unhappy unless with people. 

Under guidance of this trait new and effective expressions may 
be found to satisfy the craving for social intercourse. Habits no 
longer dominate the trait; rather it is the trait that forces the forma- 
tion of new habits, congenial and serviceable to the trait. The trans- 
formation of motives from the simple conditioned responses of in- 


2 93 

fancy is complete. The trait has transcended its specific foci of origin. 
Neither conditioned response, nor specific habit, nor identical ele- 
ments, nor instinct represents the condition that prevails. Sociability 
has become a deep and characteristic quality of this individual’s per- 
sonality. Its expression is variable; a wide range of equivalent stimuli 
arouse it. Furthermore its structure has changed with time; for not 
only has it become a pervading style of behavior, but also a motiva- 
tional system basic in the structure of this personality. The trait has 
become autonomous. 

Trait and Attitude. A trait is a form of readiness for response; 
so too is an attitude. A trait is individualized, distinctive of its pos- 
sessor; so too may be an attitude. A trait guides the course of be- 
havior, and may often become dynamic and compulsive as well; so 
may the attitude. Both may be regarded as biophysical in nature, 
combining, in any proportion, the fruits of heredity and the fruits 
of learning. Are trait and attitude, therefore, equivalent concepts? 

There is indeed a great similarity, and there are instances where 
it is a matter of indifference whether a certain disposition is called 
an attitude or a trait. To take one example, introversion and extro- 
version have been regarded both as traits of personality and as atti- 
tudes toward reality. Either designation is acceptable. 

1. Yet there are three distinctions. In the first place an attitude has 
a well-defined object of reference, either material or conceptual; 
whereas traits have no such definite reference to objects. One’s point 
of view toward liquor or frog’s legs, toward arctic exploration, di- 
vorce, or Fascism are attitudes, but one’s conservative, radical, 
ascetic, indulgent, reserved, or expansive manner of behaving is a 
trait. The more numerous the objects that arouse an attitude, the 
more closely does the attitude resemble a trait. The more an attitude 
is specific and stimulus-bound the less does it resemble a trait. 

2. As this last statement implies, attitudes may be specific as well 
as general; whereas a trait may be only general. According to psycho- 
logical usage an attitude may be a narrowly limited state of readi- 
ness for response. The determindierende Tendenz, the Bevousst- 
seinslage, the Aufgabe, as sensori-motor sets of the moment, are all 
specific attitudes. They are not traits. On the other hand, in the case 
of a more widely extended attitude, e.g., toward the world at large 
(Weltanschauung), there is no real distinction between attitude and 

3. The term attitude, furthermore, usually signifies the acceptance 



or rejection, of the object or concept of value to which it is related. 
Ordinarily attitudes are favorable or unfavorable, well-disposed 01 
ill-disposed; they lead one to approach or withdraw, to affirm or to 
negate. Traits as a rule have no such clear-cut direction. They are 
often merely stylistic, and their significance is often adverbial rather 
than prepositional. But here again there are cases where the terms 
are interchangeable. A well-integrated trait of patriotism, for exam- 
ple, might equally well be a highly generalized favorable attitude 
toward all objects and values subsumed under the individual’s con- 
ception of nationhood. In such a case the only basis of preference 
is the context of the discussion. 

Both attitude and trait are indispensable concepts. Between them 
they cover virtually every type of disposition with which the psy- 
chology of personality concerns itself. Ordinarily attitude should be 
employed when the disposition is bound to an object or value, that 
is to say, when it is aroused by a well-defined class of stimuli, and 
when the individual feels toward these stimuli a definite attraction 
or repulsion. In some cases either of the terms (trait or attitude) is 
correct, as in the case of extroversion or patriotism, previously men- 
tioned, or conservatism or radicalism. If in the last two cases the 
object or value against which the person is rebelling, or which he is 
intent on conserving, can be specified, the term attitude is preferable. 
If, on the other hand, the radicalism or the conservatism is chronic 
and “temperamental,” expressed in almost any sphere of the person’s 
behavior, then the term trait fits the situation better. Narrow or 
specific attitudes are never traits. A man is fond of his dog: he has a 
kindly attitude toward it. But if in general he is thoughtful of, and 
sympathetic toward men and beasts, he has a trait of kindliness. The 
more generalized an attitude (the more difficult it is to specify its 
object or its polarity of affect), the more does it resemble a trait . 10 

Traits and Other Forms of Readiness . It is unnecessary to repeat 
here the distinctions drawn in earlier chapters between traits and 
factors, needs, instincts, and the array of both nomothetic and practi- 
cal elements proposed by various authors as the basic units of per- 
sonality. Nor need we say more about the opposition of traits to 
specific habits or to identical elements. All the demarcations have 

10 This comparison between attitude and trait is condensed from a more com- 
plete account of the significance of attitudes for social psychology and the psy- 
chology of personality: G. W. Allport, “Attitudes,” in A Handbook of Social Psy- 
chology (ed. by C. C. Murchison), 1935, chap. xvii. 


now been drawn. We are left with a concept of trait as a generalized 
and focalized neuropsychic system {peculiar to the individual ), with 
the capacity to rmder many stimuli functionally equivalent , and tc 
initiate and guide consistent {equivalent) forms of adaptive and ex- 
pressive behavior . 

This conception is not altogether novel. As previously pointed 
out, It has many approximate counterparts in psychological theory. 
Some of the dispositions listed below are virtually identical with 
traits, some represent special sub-classes of traits, and others are 
adumbrations of traits, proposed by authors having a point of view 
at least partially like the one here presented. r 

Psychological Concepts Equivalent to, Subordinate to, or Partially 
Overlapping the Concept of Traits 



directional tendency 
ego-system (Koffka) 

Eigenschaft (Baumgarten, Stern) 
foci of development 
general attitude 
generalized habit (Dewey) 

inner-personal region (Lewin) 

lineament (Boven) 
mode of adaptation 

mode of adjustment 
motor-perceptual region (Lewin) 
need integrate (Murray) 

Neigung (Lazurski) 

Richtungsdisposition (Stern) 
Rustungsdisposition (Stern) 
sentiment (McDougall) 
style of life (Adler) 
subjective value (Spranger) 

Triebfeder (Klages) 


Again we refer to the sharp contrast between the theory of traits 
and the doctrine (any doctrine) of types. Unlike traits, types always 
have biosocial reference. A man can be said to have a trait; but he 
cannot be said to have a type. Rather he fits a type. This bit of usage 
betrays the important fact that types exist not in people or in nature, 
but rather in the eye of the observer. Type includes more than is in 
the individual. Traits, on the contrary, are considered wholly within 
the compass of the individual. The crux of the distinction is that in 
type the reference point is always some attribute, or cluster of corre- 
lating attributes abstracted from various personalities, a biosocial 


reference defined by the interest of the particular investigator (cf, 
Figure 2, p. 15). 

Many kinds of typology flourish. There are literary types, energy 
types, pathological types, constitutional types, eidetic types, statistical 
types, and ideal types. Whatever the kind, a typology is always a 
device for exalting its author’s special interest at the expense of the 
individuality of the life which he ruthlessly dismembers. Every 
typology is based on the abstraction of some segment from the total 
personality, and the forcing of this segment to unnatural promi- 
nence. All typologies place boundaries where boundaries do not 
belong. They are artificial categories. 

This harsh judgment is unavoidable in the face of the conflicting 
claims of various typologies. Many of them pretend to embrace the 
total personality, and to follow the cleavages that occur in nature. 
But the very typologies that have proclaimed themselves “basic,” 
contradict one another. Compare, for example, the supposedly foun- 
dational types of Kretschmer, Spranger and Jaensch. Certainly not 
one of these typologies, so diverse in conception and scope, can be 
considered final, for none of them overlaps any other. Each theorist 
slices nature in any way he chooses, and finds only his own cuttings 
worthy of admiration. 

Glance at the popular dichotomous types: extrovert— introvert, 
tough-minded— tender-minded, sensory— motor, masculine— feminine, 
subjective— objective, Apollonian— Dionysian, Philistine— Bohemian, 
sthenic— asthenic, and the like. What has happened to the individ- 
ual? He is tossed from type to type, landing sometimes in one 
compartment and sometimes in another, often in none at all. The 
entire approach is external, directed toward abstracted points of 
similarity among men, rather than toward the integral neuropsychic 
make-up of any one individual man. 

The present vogue of typology comes largely from the influence 
of psychiatry, which, for a long time following Kraepelin, had great 
interest in the classification of mental disease. Today, just at the time 
when many psychiatrists are turning away from classifications to the 
study of mental disturbance in the individual case, this discredited 
approach is sifting through to plague the psychology of normal per- 
sonality. We may now, it is said, classify a person as cycloid, schizoid, 
melancholoid, manoid, paranoid, hypochondroid, hysteroid, imbe- 
ciloid, neurasthenoid, or epileptoid. And so we may, if we are inter- 
ested in fitting him to the categories of mental disease. Quite as 



properly the same person may be classified as of the sensory or motor 
type, if we are interested in his style of reaction; or of the visual, 
kinesthetic, or auditory type, if we are interested in his imagery; 
Nordic, Alpine or Mediterranean, if we are interested in his race. 
But if, as psychologists, we are interested in him as an individual, 
his multiple memberships in these miscellaneous types are all seen to 
be factitious. Better to study him alive, to observe his own individual 
traits in action, and see what they signify in his own life, even if by 
this method we cannot always neatly place him in our favorite filing 
cabinet. If the information to be gained indirectly through fitting 
him to one typology or another is worth anything at all, it will be 
sure to appear, much more significantly, when his traits are analyzed. 


Strictly speaking, no two persons ever have precisely the same 
trait. Though each of two men may be aggressive (or esthetic ), the 
style and range of the aggression (or estheticism) in each case is 
noticeably different. What else could be expected in view of the 
unique hereditary endowment, the different developmental history, 
and the never-repeated external influences that determine each per- 
sonality? The end product of unique determination can never be 
anything but unique. 

This evident fact is one that most psychologists have great diffi- 
culty in accepting. If individuals cannot be compared with one 
another in respect to the same traits, what is to become of the psy- 
chology of personality as a “scientific” (i.e., nomothetic) discipline? 
Outraged at the prospect, one psychologist exclaimed, “I think it is 
nonsense to say that no two men ever have the same trait. I mean, 
of course it is true, but it is one of those truths that can’t be ac- 
cepted.” The die-hard nomothetist feels that in sheer loyalty to 
science, he must search for nothing but common and basic variables, 
however great the resulting distortion of the individual structure. 
But if he will put aside those methodological fetishes that make mani- 
fest truths unacceptable to him, and adopt a more liberal conception 
of his science, he will see that the dilemma is not fatal. The case 
for the ultimate individuality of every trait is indeed invincible, but 
there is nevertheless a certain logic that justifies his search for com- 
parable and mensurable units. 

For all their ultimate differences, normal persons within a given 

29 g the theory of traits 

culture-area, tend to develop a limited number of roughly comparable 
modes of adjustment. The original endowment of most human beings, 
their stages of growth, and the demands of their particular society, 
are sufficiently standard and comparable to lead to some basic modes 
of adjustment that from individual to individual are approximately 
the same. To take an example: the nature of the struggle for survival 
in a competitive society tends to force every individual to seek his 
own most suitable level of aggression . As the saying goes, everyone 
must be either a boot or a door-mat. One child as he matures finds 
that a constant effort to dominate his fellows is for him the most 
successful design for living; another finds that for him there is more 
satisfaction in a characteristic yielding or submission. Somewhere 
between the extremes of exaggerated domination and complete 
passivity, there lies for each normal individual a level of adaptation 
that fits his intimate requirements. The psychologist does well to 
recognize all these possible gradations and to postulate a common 
variable (in this case, ascendance— submission) which, though rough 
and approximate, permits quantitative scaling. He does not measure 
directly the full-bodied individual trait that alone exists as a neuro- 
psychic disposition and as the one irreducible unit of personality. 
What he does is to measure a common aspect of this trait, such a 
portion thereof as takes common cultural forms of expression and 
signifies essentially the same manner of adjusting within the social 

To make the case quite concrete, let us suppose that the investigator 
wishes a scale for the purpose of comparing individuals in respect to 
the common (continuum) trait, ascendance— submission, mentioned 
above. He recognizes that he is concerned only with an aspect of 
neuropsychic dispositions that differ in each person. (There are end- 
less varieties of leaders, dominators, aggressors, followers, yielders, and 
timid souls). What he does is to shut his eyes to the uniqueness of 
each case, and then seek a uniform schedule of test items that will 
force each individual into the same continuum. He selects plausible 
items from common cultural situations , 11 determines their diagnostic 
significance for the dimension he has in mind, standardizes and vali- 
dates his scale as a whole, and emerges at last with a “personality test.” 

Measured in this way, ascendance— submission is not, strictly speak- 
ing, a trait at all. It is rather a bipolar directional scale or common 
continuum upon which a certain common aspect of true individual 
traits is measured. 

11 a selection of such aspective” items is given on pp. 412 f. 



In the strict sense of the definition of traits (p. 295) only the 
individual trait is a true trait: (a) because traits are always in indi- 
viduals and not in the community at large, and (b) because they 
develop and generalize into dynamic dispositions in unique ways 
according to the experiences of each individual. The common (con- 
tinuum) trait is not a true trait at all, but is merely a measurable aspect 
of complex individual traits. 12 

The question naturally arises whether common traits should be 
called traits at all. At first thought it seems confusing to speak of both 
biophysical dispositions (true traits) and empirical continua (products 
of abstractive analysis) by the same name. And yet for three reasons 
it seems best to let the term stand for both the individual and the com- 
mon dimension. 

1. In the first place, the term has a widely established generic usage. 
Hitherto it has not meant merely an individual trait (as in the defini- 
tion on p. 295). In psychological parlance empirical continua are 
ordinarily called “trait scales”; and this usage must be respected. It is 
merely a step toward precision to speak of these continua as “common 
trait scales.” To do so requires no radical change in usage. 

2. The concepts of individual trait and common trait are comple- 
mentary in the study of personality. What is unique and what is uni-, 
versal both need to be explored. Though the former approach is the 
more fundamental, the latter constitutes a large chapter in research 
(cf. Chapter XV). Hence, by preserving the term trait with its proper 
qualifying adjectives (“common” or “individual”), the two major 
methods of approaching personality are simultaneously covered, and 
a useful generic concept results. The psychology of personality is, in 
this sense, the scientific study of human traits. 

3. There is an even stronger argument in favor of the epithet “com- 
mon trait.” Although it is undeniable that no two people have pre- 
cisely the same trait, yet there are certain aspects of personality in re- 
spect to which all people in a given culture may reasonably be com- 
pared. It was shown above that ascendance and submission, to take 
two typical common traits, represent forms of adjustment which all 
individuals by virtue of biological necessity and cultural pressure 
must in varying degrees adopt. The same may be said for many other 
common forms of adjustment: gregariousness, talkativeness, tactful- 
ness, radicalism, money-mindedness. Every person is also a socius. He 

12 This logic of traits agrees in all essentials with the position defined by F. H. 
Allport, “Teleonomic Description in the Study of Personality,” Char. & Pers., 1937, 
6, 202-214. There is however a difference of terminology. This author means by 
“trait” what is here called a common trait, and for individual trait he proposes the 
designation “teleonomic trend” 


must come to terms, though in varying degrees, with the demands of 
his culture. His basic traits are always his own, but they have a social 
aspect that is easily separated for analysis. Common traits, then, are 
not wholly arbitrary variables. They rest on an evolutionary and cul- 
tural logic. Common traits are those aspects of personality in respect 
to --which most mature people within a given culture can be compared . 13 

The trap to be avoided is the erroneous assumption that the 
common trait ever corresponds exactly to the neuropsychic disposi- 
tions of individuals. Perhaps in occasional cases there may be a close 
accidental correspondence; but it is always precarious to measure 
common traits on the assumption that they are direct measurements 
of personal dispositions. A test has different meanings to different 
subjects, and their responses may have quite different significance. 
Time and again subjects in submitting themselves to a personality 
scale complain that the test does not fit them. They seem to sense tha t 
the scale misses their individual traits. There is no convincing reply 
to these critics. The best the psychologist can do is to assure them 
that the test is only a “rough measure,” and that in individual cases 
it may indeed fall wide of its mark. 

Since common traits are at best only convenient approximations, 
the psychologist should employ whatever technical aids he can to 
make his scale as serviceable as possible. By using statistical formulae 
he may determine the reliability of the scale; he may standardize it 
for various classes of people, malting it more appropriate for voca- 
tional guidance or other practical uses. If by statistical analysis he 
finds his scale to be unreliable, he may discard the items found not 
to correlate with the remainder, and in so doing, of course, he alters 
to some extent his original conception of the common trait in ques- 
tion. In working out his scale he thus combines an initial conception 
of the trait with various empirical checks, and emerges with a some- 
what mixed product that often gains through its applicability to a 
whole population of subjects what it loses in sensitiveness for the 
individual case. 

In this process, factor-analysis is one of the statistical devices that 

_ 18 At a higher level the distinction between the individual trait and the common 
trait is reflected in the distinction between the person and the socius. The former 
is the complete man, a unique biophysical product in whom cultural influences have 
been embedded in individual ways within a biological ground. The socius is the man 
viewed in reference to his social status. His beliefs, attitudes and traits are regarded 
as conforming to, or deviating from, societal standards. The frame of reference in 
the former case is the person himself; in the latter case, external social norms. 



may assist in fashioning common traits for efficient measurement. The 
scaling of “pure” and “basic” factors— while unjustified from the 
point of view of the organic structure of personality— has undoubted 
utility for certain practical purposes. Occasionally too factor- 
analysis may succeed in cutting through certain complications of a 
priori conception, and shed new light upon problems of defining the 
common trait in question. 14 But one must be cautious lest the last 
state be worse than the first, for some factorial products are psycho- 
logically meaningless. Technical aids should never be allowed to 
lower the basic requirement of psychological intelligibility. 

A nice illustration of the difference between common and indi- 
vidual traits is offered by Conrad’s study of ratings. Three teachers 
were required to rate a number of children of preschool age upon 
2 3 1 common traits, thus being forced to make the assumption that 
all the children did possess exactly these selfsame qualities in some 
degree. Proceeding on such a false assumption there was only a low 
agreement among the teachers, ranging from +.14 to +.78, with 
a median of +.48. Many of the children, it seems, were given ratings 
on the basis of sheer guesses, because the investigation required that 
each child receive a rating on every quality. But in the course of 
the same study, the teachers were instructed to star their ratings on 
such qualities as they considered to be of “central or dominating im- 
portance in the child’s personality.” On this part of their task the 
teachers agreed very well in their judgments: their ratings on such 
starred qualities correlated from +.93 to +.96. 15 This result shows 
that the low reliability of rating may often be due to the fact that 
the subjects are forced into a scale where they do not belong. In a 
few cases (the starred qualities) the common trait concept seemed 
to correspond fairly well to some striking individual traits, but in 
most cases the common designations fell wide of the mark. 

Consistent with its nomothetic tradition, psychology up till now 

14 An instance is Guilford’s reduction of introversion to certain component 
common traits. Two of these components (the principal factors) are social 'with- 
drawing and emotionality . Each is a meaningful trait, undoubtedly susceptible of 
measurement. In this case a concept that was troublesome to psychologists receives 
clarification. Jung’s initial logic in respect to introversion, provocative though it was, 
turned out, at least for the purposes of measurement, to be amenable to improve- 
ment. Two, or perhaps more, separate scales, one for social withdrawing and one 
for emotionality, should probably replace the coarser omnibus scales hitherto relied 
upon to measure the somewhat amorphous common variable. (J. P. and R. B. Guil- 
ford, /. Abnorm. & Soc . Psychol 1934, *8,. 377-399). 

15 H. S. Conrad, /. Educ. Psychol, 1932, 23, 671-680. 



has been more interested in common than in individual traits. (The 
clinician and the therapeutist are exceptions to this rale.) Yet 
nothing is more essential in the entire field of personality than an 
adequate recognition of individual traits. The methods adapted to 
the study of common variables do not readily transfer to the study 
of individual traits. New techniques are needed. No investigator, for 
example, would undertake to scale paranoia in the general popula- 
tion; yet paranoia may on occasion be the very core of personality. 
Nor would he attempt to scale -fastidious exhibitionism , a trait for 
which Beau Brummell was famous. Goethe, who believed that there 
is always a sense in which everything is true, had a charming trait 
of listening to everyone without contradicting him. This is not a 
common enough trait to be scaled, but it is important in the under- 
standing of Goethe’s personality. Nor is its antithesis common— the 
negativism of those, who, like Benvenuto Cellini, or Samuel Johnson, 
are notoriously contredisant . When this uncommon trait does exist, it 
is exceedingly important, for every proposal, every assertion, almost 
every word, arouses a rejection. Dickens immortalizes the trait in Mr. 
Grimwig; but Grim wigs are rarely encountered. 

There are many other characters, both in fiction and in history, 
known for a single outstanding trait: Uriah Heep for his sycophancy, 
Rose Dartle for her peculiar insinuations, Oblomov for his pro- 
crastination, Mrs. Jellyby for her presbyoptic philanthropy, Micaw- 
ber for his empty optimism, Chesterfield for his self-conscious good 
breeding, the Marquis de Sade for his sexual cruelty. Our vocabu- 
lary has been enriched by the naming of individual traits after these 
celebrities. What better proof could there be for the existence of 
cardinal traits too rare to be measured as common traits in a general 

A Partial List of Trait-designations Derived from the Names of 
Historical or Fictional Characters 

Beau Brummell 












Don Juan 





John Bull 



















Tolstoy ian 


The orthodox (nomothetic) psychologist will ask, “But what can 
psychology do with such individual traits? They cannot be measured; 
and they defy man-to-man comparison.” Actually a great deal can be 
done. Individual traits, no less than common traits, are susceptible of 
genetic, analytical, and even experimental study, in the laboratory 
and elsewhere. 16 Almost any method of studying personality (cf. 
Chapter XIV), other than tests and common scales, is adapted to the 
study of individual traits. After all, the comparison of individuals is 
only one of the goals of the psychology of personality. Understand- 
ing the individual case and determining the laws of the individual’s 
development are just as legitimate and even more important goals. 


There are approximately 18,000 terms (chiefly adjectives) in the 
English language designating distinctive and personal forms of be- 

16 Perhaps they may sometime, in a rather novel sense, be susceptible likewise 
of measurement. F. H. Allport suggests the possibility of developing an intra-indi- 
vidual continuum for each personality separately, in respect to which, one leading 
trait may be measured for its intensity and self-consistency without any reference 
whatsoever to other individuals. 

The following instance suggests the approach: “There was a certain boy whose 
behavior at school was reported by his teacher as exemplary, while his home be- 
havior was a cause of grave concern to his parents. At school he was orderly, indus- 
trious, and attentive, while at home he was noisy, unruly, and a bully toward the 
younger children. Rated upon a societally standardized continuum of degrees of a 
trait, e.g., the trait of ‘tractability,’ we find the results immediately ambiguous. The 
intra-individual distribution for both spheres (home and school) together becomes 
bi-modal rather than normal. The [common] trait approach breaks down. When, 
however, we describe his behavior teleonomically, that is, in terms of what he is 
really trying (and we do not mean consciously trying) to do, we may find a basic 
consistency underlying the contradiction in these two fields of his behavior. For 
example, the boy might, in both lines of conduct, be acting to gain the attention 
of his elders. An hypothesis concerning a possible law of this individual’s general 
behavior is thus provided, which can be subjected to the test of verification through 
wider sampling and measurement.” (Char. & Pers., Duke University Press, 1937, 5, 
pp. 206 f.) 

The proposal is that every instance of behavior may be studied for its closeness 
to a hypothetical central trait in this boy’s life, viz., “trying to gain attention of 
his elders.” If many acts clearly fall within this teleonomic continuum, then the 
trait is no longer hypothetical but empirically established, and its range and in- 
tensity may be statistically determined. 



havior . 17 At first sight this vast array of verbal symbols seems chaotic 
and wholly outside the psychologist’s field of interest. But the more 
these terms are studied, the more instructive and pertinent do they 

There is no denying that trait-names bear a very complex relation 
to the underlying structural units of personality. As a first step in 
clarifying this relation let us consider their origin. They seem to have 
come into existence in response to two entirely different human 

In the first place men experience a desire to represent by name such 
mental processes or dispositions of their fellows as can be determined 
by observation or by inference. There is a demand for depicting per- 
sonality as accurately and as faithfully as possible, for with a suitable 
term, corresponding to authentic psychological dispositions, the abil- 
ity to understand and to control one’s fellows is greatly enhanced. 
There is then reason to suppose that trait-names are not entirely ar- 
bitrary, that they are to some extent self-correcting, for there is little 
to gain by preserving through names an erroneous belief in merely 
fictitious or fabulous entities; there is everything to gain by using 
terms that designate true psychic structures. If this consideration were 
the only basis underlying our vocabulary of trait-names we shoulc 
find the correspondence between linguistic convention and psycho- 
logical truth very close, much closer than it actually is. 

There is, however, a second influence determining our lexicon of 
trait-names, namely, the tendency of each social epoch to characterize 
human qualities in the light of standards and interests peculiar to the 
times. Historically, the introduction of trait-names can be seen to fol- 
low this principle of cultural (not psychological) determination to a 
striking degree. 

Astrological superstition produced lunatic , jovial, saturnine , and 
mercurial Galenian medicine which prevailed in England until the 
time of Harvey brought the term temperament, and with it quite nat- 
urally, sanguine , choleric , melancholic , phlegmatic , good-humored , 
bad-humored , as well as cold-blooded , hearty , heartless , and cordial 
(derived from the belief that the heart was the seat of the intellect 
and feeling). Following the Protestant Reformation came some of om 

17 Cf. G. W. Allport and H. S. Odbert: “Trait-Names: a Psycho-lexical Study,” 
Psychol Monog 1936, No. 21 1. This monograph was originally designed as an 
appendix to the present volume, but its length made separate publication necessary. 
The passages reprinted here represent an abbreviated summary of the study, but 
the original should be consulted for a full statement of issues insufficiently developed 
here, as well as for the complete list of 17,953 terms. 



most indispensable trait-adjectives, reflecting the introsp dveness of 
the period, among them, sincere , p/tfzzi-, bigoted , precise , ianatic, also 
the substantives, self-regard , self-assurance , self-love , self-confidence , 
and self-esteem. Selfish is a term coined by the Presbyterians about 
1640. To the aristocratic seventeenth century belong famous , callous . 
countrified , disingenuous . Political upheavals are responsible for such 
terms as Tory, democrat , and radical 

With the growing subjectivity of literature in the eighteenth cen- 
tury came numerous terms derived from self-analysis: day-dream, de- 
pression, ennui, chagrin, apathy, diffidence ; and new (more subjec- 
tive) meanings were attached to older terms, reverie, excitement , con- 
straint, embarrassment, disappointment and others. In couuly circles 
in the eighteenth century persons were described as prim, demure, 
gawky, enthusiastic, interesting , and boresome . To recent years belong 
a surprising number of new expressions, still for the most part slang: 
booster, rooter, knocker, hoodlum, climber, yes-man , four-flusher, 
crabber, cake-eater, cloiseler , gigolo , flapper , racketeer ^ abbitt . To z/;/i 
ever-increasing vocabulary of human characteristics ^ ychology has 
contributed its share: introverted, extroverted, neurotic , regressive, 
psychasthenic, eidetic, cyclothymic, schizoid , and the like. 

It is therefore certain that trait-names are not univocal symbols cor- 
responding throughout the ages to fixed varieties of human disposi- 
tions. In spite of the fact that the names probably would not have 
been invented unless there were something “really there” in the psy- 
chic make-up of individuals to call forth the new’ designations, still 
the symbols themselves are changeable and elusive. They are invented 
in accordance with current cultural interests; their meaning often 
varies, and some fall rapidly into disuse. (Although some trait-names 
become extinct, the tendency, in English at least, is for a rapid multi- 
plication of terms designating human qualities, a reflection no doubt 
of the ever-rising interest in psychological problems.) 18 

The list of terms here under discussion is based on the 400,000 
separate terms and derivatives included in Webster’s New Interna- 
tional Dictionary (edition of 1925). The exact number in the list is 
17,953 words (obsolete terms excluded), or four and one-half per 
cent of the total English vocabulary. 

The criterion for inclusion consists in the capacity of any term co . 
distinguish the behavior of one human being from that of another. 
Terms representing common (non-distinctive) behavior are excluded, 
e.g., walking and digesting, whereas more differentiating and stylistic 

1* Q. W. Allport and H. S. Odbert, op. cit., pp. 1-3. 

3 o6 


terms applied to these same activities, such as mincing and dyspeptic , 
are included. In many cases the application of this criterion involved 
a considerable degree of arbitrariness. In deciding doubtful cases the 
dictionary definition was followed: if in any of its meanings a term 
might be differentially employed in characterizing personal behavior 
it was admitted. 

Adjectival and participial forms have been preferred throughout; 
nouns and adverbs appear only where no corresponding adjective or 
participle exists, or else in cases where their meaning is distinctive 
{e.g., both Quaker and Quakerish are included). 19 

Having defined certain principles for the selection of terms, the 
next task is to determine whether a basis can be found for a psycho- 
logically significant classification of the terms. A division of the list 
into four parallel columns seems to fill the need. 

Column 1 . In this column appear those names that symbolize most 
clearly “real” traits of personality. They designate generalized and 
personalized determining tendencies— consistent and stable modes of an 
individual’s adjustment to his environment. Obvious examples are 
aggressive, introverted, sociable. These terms do not imply merely 
temporary ai^d specific behavior as do the terms in Column II (see 
p. 309) ; they are more neutral and less censorial than those in Column 
III; and they are less metaphorical and remote in their applicability to 
personality than those in Column IV. On the other hand, since the 
decision is often arbitrary, the investigator using the list is advised not 
to depend upon Column I alone, but to consult the parallel columns 
■for added terms according to his needs and interests. The intention of 
this first column is to provide merely a minimum list of trait-names 
and not a final list. The number of terms in this column is 4,504, or 
25 per cent of the total list. 

Column II. This column contains terms descriptive of present activ- 
ity, temporary states of mind, and mood. The criterion for inclusion 
reads as follows: “Might the quality in question characterize a person’s 
mood, emotion, present attitude, or present activity (but not his en- 
during and recurring modes of adjustment)?” Typical terms in this 
column are abashed, gibbering, rejoicing, frantic. The majority of 
these terms are present participles, derived from verbs signifying dif- 
ferentiative behavior. This column' contains 4,54 s words, about 25 per 
cent of the entire list. 

Column 111 . This list is the longest of the four, and contains charac- 
terial evaluations. Typical examples are insignificant , acceptable, 

18 G. W. Allport and H. S. Odbert, op. cit., p. 24. 



worthy . The paradigm for inclusion reads “Might one judge a man as 
{worthy) without the man possessing a corresponding biophysical 
trait which may be symbolized with the same name?” It is obviously 
Impossible to think of worthiness as resident in the structure of per- 
sonality itself; it is altogether a social judgment. In this respect it dif- 
fers decidedly from benevolence , tolerance , or patience . A person 
with three such biophysical traits would no doubt be judged as 
worthy , but he never could have a neuropsychic disposition of 
“worthiness.” Some terms in this column imply no profound moral 
judgment but rather a social effect upon the emotions or moods of 
another, e.g., dazzling , irritating . These terms presuppose some traits 
in a man, but in themselves they are value-estimates and do not sym- 
bolize the psychological dispositions in him that cause him to have a 
dazzling or irritating effect upon others. This column contains 5,226 
terms, or 29 per cent of the total list. 

Strictly speaking, in the sense of the present volume, Column III is 
not a list of trait-names at all. It is, however, included for good rea- 
sons. In the first place, there are writers who consider personality to 
be essentially the social influence of an individual, and from this point 
of view these terms become especially significant; they represent not 
the reactions of the individual in question, but rather his “social stim- 
ulus-value.” Furthermore, this vocabulary of social impressions and 
characterial judgment has a certain intrinsic interest for social psy- 
chology, sociology, and ethics. 

Column IV . There are many terms of possible value in character- 
izing personality, even though they have no certain place in the first 
three columns. Since in one way or another they contribute to the 
total vocabulary of useful terms they are included in this miscellane- 
ous column. More skillful editing might have made possible the assign- 
ment of some of these words to the first three columns. Subdivision 
is also possible. One sub-group might contain terms explanatory of 
behavior, past participles for the most part (e.g., pampered , crazed , 
malformed ). Another sub-group could be made of physical qualities 
which are commonly considered to be associated directly or indirectly 
with psychological traits, e.g., roly-poly, lean , red-headed , hoarse . Still 
another group could be made of capacities or talents, such as able, 
gifted , prolific . Then there are many terms of allegorical and doubtful 
application to human personality, and still others that for various rea- 
sons are the despair of the editors. In all, this miscellaneous column 
contains 3,682 words, or about 21 per cent of the total list. 20 

20 Ibid., pp. 25-27. 


Such a classification is of necessity approximate and to some extent 
arbitrary. Experiment proves, however, that different judges agree 
fairly well upon the use of these classificatory groups . 21 The most 
significant feature of the method is the separation of neutral trait- 
names in Column I from the evaluative or censorial terms in Column 
HI. The psychology of personality must be kept free from confusion 
with the problems of evaluation (character). But the assignment of 
any given term to one of these categories rather than to the other is 
not always an easy matter. 

Concepts originating in social judgment, e.g., honest, unselfish, law- 
abiding, may and often do become ideals or guiding principles 
adopted by individuals. In this sense, the introception of an ethical 
ideal into subjective attitude turns a characterial designation into a 
true trait-name. The plan followed in the classification is to place such 
terms in Column I if it seems that the social ideal does with fair fre- 
quency become a personal ideal and become thereby a true trait of 
personality. But it is obvious that certain normative concepts, like fine, 
crazy, or perfect are too general or too unpsychological ever to cor- 
respond precisely to any veridical personal trait. 

In spite of our efforts to locate only neutral terms in Column I some 
of the terms appearing there do seem to imply censorial judgment. In 
America to say that John is self-assured, inventive, or decisive is to 
praise him; in some societies he would stand condemned. But in such 
cases as these it is clear that some definite psychological trait is the 
object of reference however much the flavor of judgment may cling 
to the trait-name employed . 32 

Such a list is not only a thesaurus of terms, but also of problems. 
Each word is a record of common sense observation regarding human 
behavior. As such, each term constitutes an authentic problem for 
psychology. With this list at hand, the investigator will not be too 
easily beguiled into an over-simplified theory of personality. 

The list here presented is drawn from the full classification, and 
presents only an illustrative sample of terms. If the reader will note 
the “flavor” of the terms in each column, he will see why only 
Column I can be regarded as a list of names of traits in the strict 
sense in which the term is used in this volume. 

21 G. W. Allport and H. S. Odbert, op. cit., pp. 54-36. 

™lbid., pp. *8f. 


A Sampling of Terms Characterizing Personal Behavior and Personality 

Column I 

Neutral Terms Desig- 
nating Personal Traits. 


















a la militaire, F. 

alarmist, n. 















atheist, n. 





Column II 

Terms Primarily De- 
scriptive of Tempo- 
rary Moods or Activi- 
ties , 



a-tiptoe, adv. 


Column III 
Weighted Terms Con- 
veying Social or Char- 
act erial Judgments of 
Personal Conduct, or 
Designating I nflitence 
on Others. 

all-round, C. 


Column IV 
M iscellaneous : Desig- 
nations of Physique . 
Capacities , and Devel- 
opmental Conditions: 
Metaphorical ana, 
Doubtful Terms , 















a la mode, F. 




















au fait, F. 




3 10 


There remains the deeper metaphysical problem concerning the 
relation of any name to the unit-structures of nature. For centuries 
this problem has been disputed. The story is too long to be re-told 
here, but the solution suggested below is fully compatible with the 
biophysical view of traits advanced in this volume. 

The theory we present then holds that trait-names are symbols so- 
cially devised (from a mixture of ethical, cultural and psychological 
interests) for the naming and evaluation of human qualities. Some of 
these terms are obviously censorial and as such have little utility for 
the psychologist. The non-censorial terms, however, are significant, 
for their common usage establishes a presupposition that some human 
beings possess actual dispositions or traits roughly corresponding to 
these symbols. There are, however, many more traits than any list of 
single names would indicate, for we often find neologisms, phrases, 
and metaphors called upon where trait-names are insufficient. 

In scientific work no single trait-name can be accepted with assur- 
ance as applicable to a given personality until its correspondence with 
a true trait has been experimentally or clinically established. Traits 
cannot be called forth by fiat; they must be discovered. 

The use of the same trait-name applied to any two different indi- 
viduals signifies merely that the dispositions of both fall within a range 
of comparable judgments. 

Although in some respects this theory follows the position of the 
Nominalists it does not agree at all with those extremists who in deny- 
ing perfect correspondence between names and traits think they must 
also deny the very existence of traits. Traits exist in exactly the same 
sense in which any mental disposition or readiness-for-response exists. 
The naming of such intangible mental states is hazardous, but it is also 
unfortunately necessary. It would be absurd to allow the difficulties 
involved to lead us into the wholly untenable nihilistic position of 
denying mental organization and readiness altogether. 

A trait-name is a range-name. Although traits are real enough en- 
tities, trait-names are essentially blankets, covering one trait in one 
person and other (similar) traits in other people. Though perceived 
as similar and labeled identically, the trait is never, strictly speaking, 
in two different human beings exactly the same . 23 

As inadequate as common speech may be in representing the 
complex structure of personality, it is several grades more adequate 
than the mathematical symbols and neologisms that psychologists 
sometimes employ. The nature of our problem forces us to seek out, 

23 G. W. Allport and H. S. Odbert, op, cit. y pp. 20 f. 


3 ” 

to identify, dynamic mental structures and sub-structures, and to 
name them. And this is necessary even though the lexicon of any lan- 
guage is far from offering a perfect catalogue of the elements of 
mental life. To use trait-names, but to use them cautiously, is, then, 
our lot. Nor need we fear them simply because they bear the age- 
ong sanction of common sense. 



The constancy of a trait is merely an ideal or limiting con- 
eept—and for two reasons. One is that man is at no moment 
of his existence merely an adaptive, self-preserving creature; 
always there is in his behavior a spark of self-development and 
growth. For this reason his finished traits are never quite fin- 
ished. In addition, a trait is never entirely independent of the 
world outside, but stands in constant active relationship to it. 

It indicates the way in which the person reacts to the world; 
but never are the stimuli that provoke the reaction entirely the 
same, and never therefore are the various expressions of one 
and the same trait completely in agreement with one another. 

The trait is each time slightly different because it confronts 
other determining conditions; and these conditions produce 
not only a special coloring in each act that a trait arouses, but 
also can influence the trait itself in a permanent way. 

—William Stem 

In the last chapter the theory of traits was presented in broad out- 
line. For the time being many special questions and complications 
were overlooked which now require full consideration. As Stern 
points out, the ever-changing nature of traits and their close de- 
pendence upon the fluid conditions of the environment forbid a con- 
ception that is over-rigid or over-simple. 


The chief danger in the concept of trait is that, through habitual 
and careless use, it may come to stand for an assembly of separate 
and self-active faculties, thought to govern behavior all by them- 
selves, without interference. We must cast out this lazy interpreta- 
tion of the concept. For, to say the least, such psychological dei ex 
machina are more than a century out of date. The basic principle of 
behavior is its comimious ftcnn, each successive act representing a 



convergent mobilization of all energy available at the moment. No 
single trait— nor all traits together— determine behavior all by them- 
selves. The conditions of the moment are also decisive; the special 
character of the stimulus, the temporary distribution of stresses and 
tensions within the neuropsychic system, all demand a special form 
of adaptive response, perhaps never again required in precisely the 
same way. 

Only one maximally integrated activity takes place at any one 
time, and this activity is the product of a final convergent path 
wherein all available energy, though not all potential energy, is 
channelized to meet the present demand. From moment to moment 
there is a redistribution of this available energy, with the result that 
consummatory acts are ever changing and are the product of the 
interaction of all manner of determining factors, of which traits are 
only one. 

Unless full recognition is given to this continuous, variable and 
convergent character of behavior, the theory of traits will become a 
purely fanciful doctrine of “little men within the breast” possessing, 
by hypothesis, exclusive control over each and every separate activity. 

A kindly little man will be made responsible for initiating acts of 
kindness, and other homunculi will be credited with acts that are 
aggressive, vulgar, or avaricious. This, in principle, is the error of 
the older faculty psychology with its list of the “active powers of 
the mind.” It is likewise an ever present peril for any dynamic psy- 
chology that conceptualizes the “forces” presumed to initiate and 
guide behavior, whether they be nomothetic forces, such as instincts, 
drives, wishes, needs, or more individualized dispositions such as traits. 

Recognizing and admitting the danger, it can more easily be 
avoided. One precaution is the constant return to the observable 
stream of behavior , the only basic datum with which the psychology 
of personality has to work. Here the principles of continuity and re- 
distribution and convergence rule supreme. Traits as such are not 
observable in the stream of behavior. What is observable is the suc- 
cession of specific adaptive acts that follow one another in close 
array. Though traits themselves are never directly observed, they 
are of necessity inferred. For without some inference of a flexible | J 

underlying structure in personality it would be impossible to account 
for the recurrent quality of the separate observable acts. And yet to 
avoid personifying the separate traits, it is ever necessary before 
assuming their presence to return to the basic behavioral evidence, 


and to demonstrate that the inference is justified by strict adherence 
to the data . 1 

Traits, then, are discovered not by deductive reasoning, not by 
fiat, not by naming, and are themselves never directly observed. They 
are discovered in the individual life-the only place where they can 
be discovered-only through an inference (or interpretation) made 
necessary by the demonstrable consistency of the separate observable 
acts of behavior. 

There are many technical aids that may be employed in the dis- 
covery of traits, but each is nothing more than a refinement of the 
method used by anyone in everyday life. The refinements reduce the 
likelihood of error, but the procedure is basically the same. In ordi- 
nary life a judgment to the effect that a certain man is a “zealot for 
justice,” is based on observation of his conduct in situations where 
he is seen to be active and aggressive in demanding equity or in 
championing the underdog. The judgment, of course, may be mis- 
taken; it may be based on some one impressive single instance which, 
divorced from the usual pattern of circumstances, is not at all typical 
of the man’s ordinary behavior. Since common sense has no accepted 
criterion for the inference of a trait, erroneous first impressions are 
often the consequence. Though the psychologist also relies on ob- 
servation and inference, he is never satisfied with unverified first 
impressions. It is his business to demand a more exacting demonstra- 

The methods for establishing a trait depend upon the kind of 
trait that is the object of investigation, whether it be a common trait 
or an individual trait. In the latter case, the so-called clinical method 
is ordinarily used, especially by psychiatrists, psychoanalysts, consult- 
ing psychologists, and writers of case histories. The investigator 
makes a sustained study of a particular individual and on the basis of 
personal acquaintance with the case pronounces that such and such 
traits are outstanding . 2 Now, the objection to this method is that it 

1 This concession to the operational criterion for the validation of concepts 
may be made without subscribing to the entire creed of opcrationism which would 
brand as meaningless ’ every psychological hypothesis or inference not instantly 
defined by objective evidence. Traits are more readily defined in operational terms 
than are certain other concepts equally indispensable to the psychology of person- 
ality, such as self-consciousness, similarity, intention, intuition, and other forms of 
mental activity belonging to the realm of immediate experience. 

2 This is likewise the method of biography. Cf. R. B. Perry’s characterization 
of William James: “Turning to James’s benign traits, I find four that are peculiarly 
pervasive: sensibility, vivacity, humanity, and sociability.” The Thought and Charac- 
ter of William James, 1935, Vol. II, p. 682. 



rests ultimately upon the “intuition” of the investigator and is with- 
out the benefit of objective verification. The clinical method stands 
next door to common sense in its reliance on subjective pronounce- 
ments. But in defense of the clinical method it may be urged that 
prolonged critical probing of many-sided material, even though exe- 
cuted by a single mind without external checks, tends to be self- 
validating. Erroneous first impressions are transcended and the true 
pattern emerges with acquaintance. This line of defense does not, of 
course, justify all dogmatic assertions concerning traits, nor does it 
deny the inherent danger of subjective diagnosis. It merely hints at 
the fact (more fully discussed in Chapters XIV and XIX) that direct, 
synthetic judgments have their place even in scientific studies of 

Experimentation also provides evidence for the existence of indi- 
vidual traits. Whenever diverse tasks set in the laboratory are re- 
sponded to in uniform ways, whenever many stimuli and many re- 
sponses are found to be equivalent, a trait is safely inferred. 3 Rating 
is another method, especially valuable when several judges are em- 
ployed for the purpose of checking one another. The method of 
starred ratings (cf. p. 301) is particularly useful in the establishment 
of individual traits. 

Serviceable especially in studying the traits of young children is 
the method of the time-sample. Two or more independent observers 
for short periods of time watch the natural and unrestrained activi- 
ties of children. Each observation may last for no more than one 
minute. Through repetition, say, at intervals of one hour for several 
days, the accumulation of such observations gives a significant sam- 
pling of the behavior. When different observers agree on the frequent 
recurrence of certain forms of conduct an excellent basis exists for 
the inference of a trait/ 

Time-sampling may be used either for the discovery of individual 
traits, or for the discovery of common (continuum) traits. In most 
researches to date the consistency of each child within a group has 
been studied with reference to the same common variable. There is 
nothing in the method, however, to prevent its being used for indi- 

3 Such laboratory demonstrations of traits can take place, however, only when 
the tasks set for the subject are diverse enough to permit the discovery of the 
equivalence of many stimuli and of many responses. Cf. G. W. Allport, Char. & 
Pers. y 1933, 1, 259-264. 

4 Cf. the work of F. L. Goodenough on such qualities as leadership, dramatic 
play, and reticence at the nursery school age. J. Juv. Res.> 1928, 12, 230-235. 


vidual traits. A child who by intermittent observation is found always 
lost in fantasy, or vigorously active, or reading a book, may safely be 
said to have a certain trait or interest even if he is not directly com- 
pared with other children. 

To aid in the discovery of common traits a variety of statistical 
devices may be used. Perhaps the most familiar are the various 
methods of determining the reliability of tests for some one trait. 
If a test, constructed of many items, and sampling a wide field of 
activity, is found to be reliable, this fact proves that subjects reply 
to the diverse items in a consistent way. If a test, for example, de- 
signed to measure social withdrawing, is reliable, it shows that sub- 
jects who “withdraw” in one situation (e.g., who say that they keep 
a diary record of their inner life) are likely to “withdraw” in others 
(e.g., admit being embarrassed when meeting strangers). If every re- 
sponse is positively correlated with every other it shows that the 
test is internally consistent, which means in turn, that the subjects, 
by and large, respond consistently and uniformly to the scale. 

The contributions of statistical method to the demonstration of 
common traits are numerous; there are all manner of measures of 
correlation, contingency and association. Employed according to the 
needs of the special problem and the nature of the material in hand, 
each enables the investigator to determine the degree of co-variation 
of different forms of behavior. A demonstration of the persistence 
of association is always the first step in inferring the existence of a 
common trait. 6 

The procedure of assembling and examining the evidence for a trait, 
together with some of the difficulties involved, can be illustrated by 
reference to Dudycha’s study of punctuality. 0 

The data for this study consisted of over 1 5,000 observations on the 
time of arrival of some 300 students at various exercises and collegiate 
gatherings (as diversified as basketball games, eight o’clock classes, 
vespers, private conferences, and college commons). Here is typical 

5 Any measure of association, to be sure, requires interpretation. Whether or 
not a given magnitude of association proves the existence of a trait frequently de- 
pends on the bias of the investigator who interprets the measure. Time and again, 
for example, correlational coefficients between various indices (presumably of the 
same trait) are no larger than +.35 or +.45. Having in mind the low predictive 
value of such coefficients one investigator will say that no trait can be inferred; an- 
other, aware of the intrusion of unwanted variables into his experiment, or having 
supplementary evidence in mind, may consider such low coefficients indicative. 
There is no certain criterion. 

6 G. J. Dudycha, “An Objective Study of Punctuality in Relation to Personality 
and Achievement,” Arch, of Psychol., 1936, No. 204. 



raw materia! for the beginning of an extended study of traits. In this 
particular case, the problem was to determine whether punctuality 
(or earliness, or lateness) was a characteristic trait for every student. 

After transforming into standard scores the records of earliness and 
lateness for each student for every occasion on which he was ob- 
served, the usual product-moment (Pearsonian) method of correlation 
was applied. The results were indecisive. There seemed to be only a 
slight tendency for students to maintain the same position of earliness 
or lateness in arrival. Eleven of the twelve coefficients of correlation 
were positive, but the largest (between arrival at entertainments and 
at college commons) was only -I— .44. From such evidence, the author, 
leaning to the side of caution, concludes, “We can hardly say that 
there is a general trait of punctuality —or its opposite, tardiness— which 
operates in the whole group.’ 7 

Coefficients of correlation, however, have a propensity for accentu- 
ating slight variations in an individual’s behavior, thus speciously re- 
ducing the evidence for traits, which of course require no such per- 
fect correspondence of measures. The author makes this point, and 
explains his remedy in the following words: “Pearson ids do not 
clearly reveal trait behavior because the variations in that behavior are 
accentuated. Since in life situations we are not interested in slight dif- 
ferences of a few minutes, but rather in whether a person is early, on 
time, or late, fine differences in promptness or tardiness, for practical 
purposes, are of little significance. Hence we must employ such statis- 
tical tools as make use of broad categories, namely, Chi square 
which is not a measure of the degree of association , but of the prob- 
ability that association is or is not significant . It must be emphasized 
further that those variations which decrease the fs , as already noted , 
are largely within the broad categories selected and hence have no 
effect on the Chi squares. Since Chi square reveals whether there is 
consistency in a fairly broad area of behavior, and since the obtained 
values indicate a significant association, on the average, we must con- 
clude that there is evidence in support of a general trait of punctuality , 
when punctuality is not conceived of as referring only to isolated, 
specific events, but to a trait which is variable within limits” (p. 30). 
Thus does a different statistical tool lead to opposite conclusions! 

The author continues his analysis. He points out the inescapable 
weakness of any conception of a common (continuum) trait: “The 
distinguishing characteristics of individuals are very effectively hidden 
when a whole group, distributed according to measures on a certain 
trait, is dealt with. In other words, since the largest part of any group, 
normally distributed, clusters about a central tendency, and since the 
individuals who compose this part of the group are not distinctive in 



the measured trait, those individuals who constitute the extremes of 
the distribution, and who are distinctive in the measured trait, are 
completely lost sight of because they constitute a minority. A far too 
common error which results from this lack of discrimination is the 
conclusion, that since the majority lack distinctiveness, there is none 
who possesses a general trait. This pitfall is avoided by using the 
method of contrasted groups, which is being used more and more 
generally in this type of study” (p. 41). 

Using the method of contrasted groups, the picture again changes, 
this time more sharply favoring traits. Examination of cases of stu- 
dents falling within the extreme 20 per cent of earliness scores, and 
within the extreme 20 per cent of lateness scores, shows a remarkable 
consistency in their behavior. The early bird is nearly always early, 
the late comer usually late (though the evidence for a well-integrated 
trait of earliness is better than for lateness, due no doubt to the posi- 
tive emphasis that social training places upon earliness. The early bird 
is trying to be prompt in every situation, but the late comer is not try- 
ing to be late. He is merely the victim of a thousand and one distrac- 
tions. Hence it is that lateness is less well integrated than earliness). 

According to Dudycha, “The unmistakable conclusion is: early stu- 
dents differ significantly from late students in punctuality in general 
and in specific situations; further they differ in both the extent and 
the frequency of earliness and lateness. . . . Further, since the con- 
trasted groups include 40 per cent of the students observed, punctual- 
ity, or tardiness, is certainly found as a general attitude or trak in the 
behavior of at least two-fifths of the students; in the other 60 per cent 
it is less well organized, and exhibits less consistency” (p. 44). 

This study demonstrates nicely the psychologist’s procedure in 
the discovery of traits, and some of his dilemmas as well. Especially 
instructive is the finding that about two-fifths of the students in their 
own personal lives have well integrated habits of arriving early or 
late. Since time-of -arrival is essentially a matter of societal standards; 
it is interesting indeed to see that in so large a percentage of cases 
a regular habit of behavior has been formed around these standards. 

What then is Dudycha’s dimension of punctuality? Is he dealing 
with common traits or individual traits? According to the terminol- 
ogy of this volume he has studied a hypothetical common trait. He 
has proved that consistency of behavior is not well marked for 60 
per cent of his subjects, and therefore that no very reliable scale 
could be created for the measurement of earliness-lateness in the 
entire population. (It should, however, be remarked that many ex- 



istent scales for common traits prove equally ill-adapted to the ma- 
jority of cases.) As a common (continuum) trait therefore, punctu- 
ality leaves something to be desired, though it is perhaps no weaker 
than others of the same genre. 

In the course of his investigations Dudycha found that about 20 
per cent of his subjects were in one aspect of their behavior highly 
consistent and markedly similar to one another: they were always 
early. Now we cannot assume that all of the individuals in this group 
had precisely the same trait; not all the early birds were early for 
the same reason. They had, to be sure, uniform habits of earliness, 
but these habits may have “belonged” in quite different trait-patterns. 
In one student the habit of earliness may have been an adjunct to a 
trait of ambition, in others to a trait of deference, of pride, of rivalry 
or even timidity. Dudycha did not work on these more basic indi- 
vidual traits. But he does demonstrate clearly the fact that common 
cultural influences may set up habits in respect to which (by aspec- 
tive analysis) many individuals may be profitably compared. In short, 
he has defined a common (continuum) trait and has established its 
range of applicability: all people may be ordered to his continuum, 
but with certainty and practical usefulness, only the extreme 40 per 
cent of the population. Such is frequently the situation with com- 
mon traits. 



Characteristic of the nervous machinery, says neurophysiology, 
is its arrangement in levels, the more complex higher levels standing 
in the dual role of driver and restrainer to the lower simpler 
levels. 7 Therefore, since traits, on the physiological side, are undoubt- 
edly neural dispositions of complex order, they may be expected to 
show motivational, inhibitory, and selective effects . upon specific 
courses of conduct. Brief as this statement is, it is the sum and sub- 
stance of the present aid from neurophysiology. So far as it goes it 
suggests that the operation of a trait is dynamic, both in governing 
the reception of the stimulus and in directing the response. But the 
information is too meager to warrant at the present time a physiologi- 
cal account of the operation of traits. 

7 Cf. C. S. Sherrington, Mental Hygie?ie, 1923, 7, p. 13. 


Two kinds of dynamic psychology were described in Chapter 
VII. One holds that the root motives of men are to be sought in the 
structure underlying the traits of personality, that is to say, in in- 
stincts, the Id, in certain original needs, wishes, or drives. Advocates 
of these theories are unable to envision traits as possessed of driving 
power. Psychoanalysis, for example, sees traits, not as systems of 
motivation, but as symptoms within the Ego, of driving power within 
the Id. For McDougall they are also individual ornamentations of 
propensities common to all men. Even Morton Prince, who regards 
traits as “obstinately persistent, enduring characteristics of the per- 
sonality,” and recognizes their determining influence on behavior, 
is of the opinion that the energy of the trait is ultimately always 
derived from the instincts . 8 According to these views instinctive or 
impulsive action takes place under the general guidance of the “con- 
tributory habituation” of traits. Traits themselves are mere formal 
(directive) determining tendencies, styles of adaptation, but not the 
mainsprings of actions . 9 This type of theory persists in making per- 
sonal nature a mere incident in the universal pattern of human nature. 
It is not the attributes of a mind-in-particular that are dynamic, but 
those of the mind-in-general. 

The other kind of dynamic psychology, advocated at length in 
Chapter VII, breaks with the nomothetic tradition completely, and 
regards motives as personalized systems of tensions, in which the 
core of impulse is not to be divorced from the images, idea of goal, 
past experience, capacities, and style of conduct employed in obtain- 
ing the goal. The whole system is integral. If biological drive plays a 
part (thirst, hunger, sex), it does so, not as the motive, but merely 
as an irritable state of bodily tissues set within an intricate and per- 
sonalized psychophysical system. 

For example, in the case of the motives commonly called sexual, 
there is, to be sure, a common biological capacity involved, but the 
concrete functioning of this capacity in each life is very different. It 
cannot be regarded as one and the same force in all personalities. It 
is not a single concrete motive, but is a factor, traceable in many 

s M. Prince, Clinical and Experimental Studies in Personality, 1929, pp. 123, 127. 

°A variant of this point of view stresses the redintegrative character of traits. 
Whenever a stimulus, or internal need, is effective in arousing any portion of the 
determining tendency, the trait as a whole is redintegratively activated. Thus aroused 
it directs the required response into characteristic channels. The trait steers , but the 
effective motives are still thought to be emotional tensions quite independent of the 


kinds of other dispositions (cf. pp. 187-189). Even a sub-classification 
of types of sexual interest does not suffice to mark off the individual 
case. To say, for example, that a certain man or woman is homo- 
sexual is by no means to characterize his or her motivation. There 
are myriad forms of homosexuality: overt, covert, active, passive, 
compulsive, sublimated, diffuse, specific, altruistic, gentle, sadistic, 
protective, adulatory, superficial, unconscious, temporary, lasting, 
esthetic, intellectual— ultimately, as many forms as there are indi- 
viduals. The patterns are not directly comparable, for their signifi- 
cance in individual personalities is never twice the same. It is inexact 
to ascribe one underlying motivational system to all. What motivates 
each person is not some element common to all individuals, but his 
own particular pattern of tensions. We may perhaps learn something 
of the nature of this particular pattern from a study of the common 
biological capacity, but must not make the fatal mistake of assuming 
that it is the abstract capacity itself that does the motivating. 

Only individualized patterns of motives have the capacity to select 
stimuli, to control and direct segmental tensions, to initiate responses 
and to render them equivalent, in ways that are consistent with, and 
characteristic of, the person himself. 

But is the personalized trait dynamic in the sense of being self- 
active? does it in and of itself initiate behavior? Strictly speaking, 
no; any disposition must be aroused before it is dynamically active. 
According to the principles of convergence earlier described, at any 
one time energies are mobilized for purposes of adjustment as needed. 
Not every motivational system is at all times in a kinetic phase. The 
successive demands of living arouse first one region of stress, and 
then another. Only through an intricate course of stimulation and 
association are determining tendencies (including traits) raised from 
a state of potentiality to activity. Either external stimuli or segmental 
tensions of an organic order may arouse them; but such antecedent 
stimulation in itself is not the motive. 

In another sense traits do initiate behavior. It is certainly not a 
ring on the telephone nor a friend’s voice over the wire that causes 
an egotist to talk for half an hour unchecked about his latest ex- 
ploits, or a gossip to recount at great length the doings of her neigh- 
bors. In these cases the response springs from deep-seated traits of 
personality, and not from a ring on the phone followed by a friend’s 
voice over the wire. Traits may be even more self-active than these 
illustrations imply. The egotist and the gossip when alone may feel 


quite restless until they have sought out an opportunity to unburden 
themselves. They seek an excuse to talk, and put themselves in the 
way of stimuli that will release the flood. A sociable person, if he 
is quite as ready to listen as to talk, may be restless until he is among 
people. An author, a housekeeper, a public speaker, a reformer, a 
craftsman, a musician, if deprived of their favorite occupation, may 
“itch” to return to their work. In many such instances of “spon- 
taneous” motivation it seems impossible to trace the precise train of 
stimulation or association that sets up the aggravated field of stress. 
The traits appear to be self -active, at least at times when they are 
not specifically inhibited by contrary courses of conduct. 

A certain man, known to everyone as generous and open-handed, 
was preoccupied at a picnic in lighting a camp-fire. Someone sug- 
gested that for supper a half-pint of cream was needed, and that it 
could probably be obtained at a nearby farmhouse. Still preoccupied 
with his fire, the generous man reached into his pocket and handed 
over two one dollar bills. There was no reason why he rather than 
any other member of the party should have footed the bill, and cer- 
tainly no reason for handing over so much money: not even a Yankee 
farmer could conceivably charge more than one dollar for half a pint 
of cream. 

What had happened in this simple situation? Caught off-guard, the 
fire-tender fully revealed his trait of open-handedness. Normally, the 
specific demands of the occasion, and a hasty bit of mental arithmetic, 
would have brought forth a silver coin, or that lacking, one of the 
dollar bills. Being preoccupied with other duties, the half-heard sug- 
gestion crossed the ever-low threshold of his generosity. The re- 
sponse was perfectly in keeping with the disposition, though some- 
what excessive and maladaptive in terms of the demands of the occa- 
sion. The fact that he offered the money at all, and the fact that he 
offered it in extravagant amounts, show the decisive influence of the 
trait, all the more decisive because it was not specifically controlled. 

Was it the trait that initiated the behavior, or was it the semi-con- 
scious suggestion that came to his preoccupied mind? The suggestion 
of course was prior in time, and was required to throw a latent system 
into action. But the same words spoken to a niggardly or unsocial 
person might have had no effect at all Whether we say it was the 
words or the trait that initiated the behavior, there is no doubt that 
the trait played the decisive role. 

Interests, ambitions, compulsions, phobias, general attitudes, in- 
clinations, hobbies, values, tastes, predilections, and the like, are all 



traits, and are at the same time motives. Yet it is not correct to say 
that all motives are traits. The demands of some especially intense 
stimulation, such as pain, thirst, and the like, lead to immediate, and 
often reflex, adaptations not well integrated with other tendencies. 
Such motives are numerous, and quite specific, not at all resembling 
traits which are always complex and recurrent systems of stress. 

Nor is it correct to think of all traits as motives— often they seem 
to have a defining or directive influence upon conduct, without true 
motivational significance. Some traits have less to do with stress than 
with style. Except in rare instances politeness, for example, seems not 
to be a motivational trait. One does not leave the house and seek out 
other people in order to be polite to them. One may seek out others 
because one is sociable and restless without their company; having 
sought them and being now in their company one may behave 
toward them in a polite manner. Nor is a man often forceful for the 
sake of being forceful; rather he employs a forceful style of behavior 
whenever he is, for other reasons, aroused to action. Such “directive” 
traits have been classed by Troland with the “praxiograms,” cortical 
patterns that do not initiate but do regulate the character of re- 
sponse . 10 Some traits thus seem to have motivational (directional) 
significance, and some mere instrumental significance. The latter are 
primarily expressive in significance, and seem predominantly motor 
in their organization; they represent styles of behaving and, unlike 
the driving traits, are seldom involved in the profundities of emo- 
tional life. The distinction is clearly drawn by Stern who calls the 
driving traits Richtungsdispositionen, and the instrumental traits, 

Now the two factors of Richtung and Rustung, however closely 
interwoven, have nonetheless a certain independence of one another, 
and the most varied relations to one another. We are therefore com- 
pelled to distinguish between those dispositions that have a prevail- 
ingly directional character from those that are principally an imple- 
mental character. The former are purposive, they have a “tendency 
to,” the latter are capacities, and have a “potency for.” 11 

This distinction is a useful one. But it must not be overworked. 
There is no sharp line between motivational traits and stylistic traits, 
between direction and manner of expression. Very often what was 

10 L. T. Troland, Cerebration and Action (Vol. Ill of The Principles of Psycho- 
physiology), 1932, p. 321. 

11 W. Stem, Die menschliche Personlichkeit, 1923 ed., p. 83. 



originally a motivational trait (e.g., interest in mechanics) becomes a 
mere instrument of expression (a skill serviceable in earning a living), 
or what was originally instrumental (e.g., skill in seamanship) may 
become a passionate interest. This continual transformation of motives 
from the level of mechanism to the level of drives, or vice versa . 
has been discussed at length in Chapter VII. 

Whether a trait is dynamic or directive is, then, a matter of de- 
gree. At one extreme are the compelling, seemingly spontaneous, un- 
restrained, obsessional traits of neurotics or psychotics . 12 Shading 
from these compulsive systems are the interests of normal people, 
which in turn vary from absorbing passions to mild predilections. 
Still less dynamic are the directive traits; nevertheless, having a cer- 
tain steering capacity, they cannot be entirely divorced from motive. 
The dynamic pressure of these expressive traits often passes un- 
noticed until an individual is compelled, through outside constraint, 
to act contrary to his usual style. The resulting discomfort and mal- 
adaptation show to what extent his expressive behavior has been 
dynamic in character. His equipment, abilities, and style, he finds, 
are not wholly matters of implementation; they too have a certain 
Drang. In brief, the distinction between driving traits and directive, 
while useful for some purposes, cannot be sustained too rigidly. 


It is obvious that what seems to be the same trait may, in different 
people, have quite diverse origins. Shyness in one person, for example, 
may be due to hereditary influences that no amount of contrary 
pressure from the environment has been able to offset; in another 
person shyness may stem from an inferiority feeling built by an ab- 
normally exacting environment. In spite of dissimilar histories, in ap- 
pearance and in effect, the shyness of these two persons may be verv 
much alike. Conversely, two youths suffering some shocking experi- 
ence of grief or bitter disappointment, objectively alike, may be 

12 It may be well to remind the reader that such traits cannot be considered 
merely as symptoms of an unconscious conflict. Miserliness, scrupulosity, compulsive 
neatness, and similar traits are more than converted expressions of frustrated eroti- 
cism, infantile or otherwise. Even in cases where such unconscious components can 
be traced, these must be regarded not as the contemporary sum and substance of 
the motive, but merely as integral with it. Whether or not some initial repression 
took place, there have been elaborations and transformations in the focal character 
of the disposition until it must now be considered as rooted in the total life rather 
than in some one recess of the unconscious. 



affected very differently. One of them becomes morose and ineffec- 
tual, lost in his trouble; the other stiffens his back and becomes more 
realistic and aggressive. The same fire that melts the butter, hardens 
the egg. 

Lev in has shown this general problem of appearance vs. under- 
lying cause to be of considerable importance in the investigation of 
personality. Descriptions in terms of here-and-now attributes are 
phenotypical; explanatory accounts, seeking underlying motives and 
stresses, are genotypical . 13 

This distinction, though valuable, has already been adequately im- 
plied in the foregoing discussion of the nature of traits. For one thing, 
common (continuum) traits are obviously conceived in phenotypical 
terms. When ascendance, perseverance, sociability, radicalism, punc- 
tuality, or neuroticism is measured for a whole population, with the 
intent of comparing all subjects in respect to the same trait, it is 
obvious that no account can be taken of the various reasons why 
different people are ascendant, persevering, sociable, or the like, nor 
can any attention be paid to the individual varieties of these traits. 
Stylistic (directive) traits also suggest Lewin’s phenotypical classifica- 
tion, since they are more closely related to external conduct than to 
root motives. Conversely, driving traits must certainly be ordered to 
Lewin’s conception of genotypical traits, for they are the very springs 
of conduct. The psychology of personality must deal with traits of 
both orders, if for no other reason than that in the course of develop- 
ment genotypes are often transformed into phenotypes, and pheno- 
types into genotypes. Chapter VII is a record of such transforma- 

Another distinction of value is that drawn by Baumgarten be- 
tween genuine traits and “pseudo-traits.” 14 This writer points out 
that erroneous inferences are often made when behavior is observed 
superficially and interpreted only at its face value. A bearer of gifts 
may not be, in spite of all appearances, a truly generous person: he 
may be trying to buy favor. A person chaste in conduct may not be 
chaste at heart. Ever so many “virtues” may be simulated and mis- 
taken by the unwary observer for true traits. “Pseudo-traits,” then, 
are errors of inference, mis judgments that come from fixing atten- 
tion solely upon appearances. The best way to avoid such errors is 
to find the genotype that underlies the conduct in question. What is 

13 K. Lewin, Gesetz und Experiment in der Psychology, 1927. 

14 F. Baumgarten, Brit. /. Psychol, 193 6, 26, 289-298. 

5 > 



the individual trying to do when he brings his gifts? If his basic 
motive is known then the sycophancy of his conduct is understood 
and the pseudo-trait gives way to the true trait in the diagnosis. 
Baumgarten’s distinction reminds us once again that the soundest 
methods and utmost of critical acumen are essential in diagnosing 
personal traits. It is only by its ability to separate true traits from 
pseudo-traits that psychology makes any advance over common sense. 


Probably nothing whatsoever in the structure of any human mind 
exists sealed up in completely independent systems. Segmental re- 
flexes seem most nearly to approach this condition, but even these 
can be markedly altered in their functioning by concurrent activity 
or tonus prevailing in other psychophysical regions. Neither are the 
oft-mentioned “dissociated complexes” and “logic tight compart- 
ments” of the mind entirely separate; their boundaries are at least 
semi- permeable. Since this is so, the influences determining conduct 
are manifold and fluid. A single act may, and usually does, result 
from the mobilization of available energy through ?nmy channels. 
Consider the task of writing a letter; it requires the convergence of 
mental sets, habits, motives of the moment, skills, stylistic traits, as 
well as the deepest of personal convictions and values. Such adaptive 
behavior always demands the effective convergence of many deter- 
mining influences, traits among others. Generalizing the illustration, 
it may safely be said that no single performance is ever a univocal 
product of any one single trait. 

At the same time, since many different adaptive acts distributed 
in time show repeatedly the same telic and expressive character, it 
becomes necessary to assume some stable and continuous influences 
at work. A markedly loquacious person, who talks at great length 
on the slightest provocation, must have some psychophysical ten- 
dency that helps to stabilize his conduct in this direction. Loquacity, 
provoked so easily (by an equivalence of stimuli), leads to fluent 
speech wherein all manner of ideas and expressions may be employed 
(equivalence of responses). Though it is not easy to conceive in 
neural terms, there must be some system with a low threshold of 
arousal, readily crossed by many stimuli and by many associational 
currents. At any one time, to be sure, the form of the loquacity is 
determined not only by the disposition in question, but by the simul- 


taneous functioning of ideational content, by special attitudes toward 
the topic under discussion, toward the interlocutor, and innumerable 
other concurrent conditions, both inner and outer. 

The term focalized disposition, previously introduced, fairly 
represents the nature of this internal press. When latent, as when 
active, the barriers between these dispositions are not rigid, for often 
an impulse arousing one trait may at the same time arouse others. 
In fact the arousal of several dispositions in varying degrees seems to 
be the rule, each contributing to the convergent conduct in propor- 
tion to ins degree of arousal. Figuratively, traits “overlap” with one 
another. They do not operate on an “all or none” principle. The 
situation is represented diagrammatically in Figure 17, p. 246, where 
interdependent traits are contrasted with independent factors. 

The trait, then, is identifiable, not by clean-cut contours or boun- 
daries, but rather by a nuclear quality, by its focus. This focus is 
essentially the telic significance of the trait, that is to say, its meaning 
to the individual as a mode of survival and mastery. The loquacious 
disposition of the talkative man is, for him, a modus vivendi. So too 
are all the other focalized (and interdependent) systems of his per- 
sonality; his esthetic interests, his thriftiness, his timidity, his neat- 
ness, his affectionate attachments, and his political conservatism— they 
are all modi vivendi. In this same sense attitudes as well as traits are 
focalized dispositions. 

By now it is clear why, either for the population at large or for 
the single individual, it is impossible, strictly speaking, to classify 
traits. Since the significant foci in any two lives are never precisely 
the same the most to be hoped for in the total population is an in- 
ventory of some of the common traits in which people may be 
roughly compared. (This procedure will be followed in Chapter XV 
where its value and its limitations will become apparent.) For the 
individual, since his traits are never wholly independent of one 
another, enumeration in terms of mutually exclusive units is not 
possible. 15 

This question of the independence of traits has important implica- 
tions for the construction and interpretation of tests. Since the tests 

is What usually happens in an attempt to list the basic traits of an acquaintance 
is that a cluster of trait-names is employed. One says of a friend, “She is one of the 
most tactful, sympathetic, and sociable people I know.” Is only one focal disposi- 
tion here intended, or two, or three? If more than one, then there is probably not 
independence among them, but some clustering tendency. Because of such clustering 
it becomes impossible to analyze a single life into wholly independent units. 


are designed for the measurement of common traits, the interpretation 
of the score of any individual subject is an awkward task. When he 
answers the questions— or performs the acts— dictated by the test, he 
is behaving in a specific adaptive manner. If the reasoning of this chap- 
ter is sound, such adaptive behavior is unlikely to be the product of 
one and only one determining disposition. The test will not measure 
one and only one trait. It is rather a rough and approximate device 
whereby the investigator hopes to tap the trait in question along with 
he knows not what other variables. There is some justification for his 
belief that the sheer length of his scale will to a certain extent lead to 
a canceling out of the unwanted and intrusive variables. Further, if 
the test has a known reliability and validity, the investigator may with 
some assurance say that whatever extraneous reasons there may be for 
a subject’s selection of responses, in the long run the trait itself prob- 
ably has the most decisive influence on the selection and on the score. 

But the problem becomes more seriously complicated when a test 
is designed (as many tests are) to measure not one trait, but several 
Diagnostic weights, different for different traits, are assigned on the 
assumption, correct in itself, that a response is determined not by one 
single trait but by many. The simple uni-trait test is contented with 
one final score, admitting that many other determinants have gone 
unmeasured . The multi-trait scale (perhaps in the interests of effi- 
ciency or of that ever-recurring naive hope of finding one solution 
for all puzzles) expects to diagnose these other determinants (so far 
as they involve additional common-traits) all at one time. 

Take, for example, Bernreuter’s Personality Inventory , an omnibus 
of 125 questions, drawn largely from three pre-existing, “logical” 
scales each of which is intended to measure one and only one com- 
mon trait. 16 With the aid of these 125 questions the scale attempts to 
measure simultaneously four common traits of personality: domi- 
nance , self-sufficiency , introversion , and neurotiewn. Each answer 
receives four scores (some of them being zero), according to its em- 
pirically (i.e.y statistically) determined diagnostic value for each of the 
four traits. Thus, one question reads, “Do you often feel just miser- 
able? ” If your answer is “?,” meaning either you don’t know how r you 
feel or else that you don’t know what the tester means, you are scored 
—3 on introversion, — 1 on dominance, and o for both neuroticism and 
self-sufficiency. Now such a response to such a question seems to bear 
a very tenuous logical relation, if any at all, to any of these four traits 
Why should a person who sets a question mark against such an item 
be judged extroverted and submissive, but neutral in respect to neu- 
roticism and self-sufficiency? 

G. Bemreuter, /. Sec, Psychol 1933, 4, 387-405. 


In order to avoid such psychological confusion, it is much more wise 
to stick to the simpler conception of a scale, wherein a single response 
is evaluated only in its diagnostic significance for one trait, deliber- 
ately leaving unmeasured all the other determining influences affecting 
the score. In so doing the temptation to depart from the precincts of 
psychological intelligibility will be lessened. 

A different attack on the dilemma created by multi-trait scales is 
to proceed in the direction of factor analysis as Flanagan has done in 
his treatment of the Bernreuter scale. 17 Bernreuter very soon found 
such a high correlation between his measures of neuroticism and intro- 
version that he could not consider them separate common traits; hence 
for practical purposes he reduced his measures from four to three. 
Flanagan goes still further; eliminating the correlations between all 
scores, he pulls out as substitutes for the four original, interdependent 
traits, two new and independent factors, which he christens self - 
confidence and sociability . After this ceremony, it turns out that if 
you “feel just miserable” you score +4 on Factor I, and o on Factor 
II. As the statistics grow better and better, the intelligibility grows 
less and less.— Contamination by statistical artifacts is not uncommon 
in scales of the multiple trait order. (From one highly sophisticated 
empirical scale comes this extreme instance: children who give the 
response word “green” to a stimulus word “grass” receive a score of 
+6 for “loyalty to the gang”— an example of empiricism gone wild.) 

Unnecessary trouble springs from assuming, as some testers do, that 
indepe?ident factors are to be preferred to inter-dependent traits 
(cf. p. 245). What if certain scales do correlate with one another? 
(Correlations of the order of +*2° to +.40 are often found.) Each 
scale may still represent a well-conceived, measurable common trait 
(if it has demonstrated reliability and validity). No harm is done by 
overlap; indeed, overlap is a reasonable expectation in view of that 
roughness of approximation which is the very nature of the entire 
procedure (also in view of the tendency of certain traits to cluster). 
Well-considered scales with some overlap are preferable to ill-con- 
sidered scales without overlap. To seek intelligible units is a better 
psychological goal than to seek independent units. 

Constructive suggestions for avoiding such difficulties as we have 
been describing will be offered in Chapter XV where the logic of 
measuring common traits will be under discussion. For the time being 
it is sufficient to say that inter-dependent traits can be measured quite 
as successfully as independent factors , and, as a rule, much more mean- 

17 J. C. Flanagan, Factor Analysis in the Study of Fersondity (Stanford Press), 




The scientific evidence for the existence of a trait always comes 
from demonstration by some acceptable method of consistency in 
behavior (the consistency being not a matter of stereotyped habits, 
but of equivalent responses) . It is simple enough to prove some people 
self-consistent: they are, for example, almost always decisive , or 
almost always fastidious. In other cases, however, the evidence is 
less conclusive. The degree of consistency that different investi- 
gators will demand before inferring a trait is a somewhat subjective 
matter. Aware of the difficulties encountered in framing pertinent 
tests, experiments, and rating scales, and aware of errors in measure- 
ment and of the unmeasurable and uncontrollable interplay of deter- 
mining tendencies in the subjects themselves, one investigator may 
be content with relatively low measures of consistency; while another, 
unmindful of, or unimpressed by, these handicaps, may demand 
almost perfect correspondence among his measures before inferring 
a trait. 

To argue against this latter position is not necessarily to lower 
standards of psychological investigation. Perfect consistency will 
never be found and must not be expected. There are many reasons 
why this is so. For one thing, in the same personality, traits often 
contradict each other. People may be both ascendant and submissive, 
perhaps submissive only towards those individuals bearing traditional 
symbols of authority and prestige; and towards everyone else ag- 
gressive and domineering. So-called guilt-behavior may betoken con- 
tradictory traits: the overt virtue may be compensatory for the 
hidden vice. Ambivalence in one’s feelings of loyalty and affection 
also makes for contradictions in traits. Every person has conflicts, 
frequently expressed in antagonistic dispositions. The ever-changing 
environment raises now one trait and now another to a state of active 

Likewise harmful to perfect consistency is the omnipresent princi- 
ple of convergence. No trait operates alone. The adaptive act of the 
moment is only partially a function of the one trait. It is determined 
as well by many other traits (even by contradictory traits, if the 
situation has dual “valence”), and by all manner of specific attitudes, 
by mood, and by momentary conditions. The excessively methodical 
person may become careless and demoralized in his haste to catch a 


train; the meticulously truthful person may lie if his life is at stake. 
Moreover, in every personality there are peculiarly specific dissoci- 
ated habits. A person, otherwise polite and considerate, may, because 
of special prejudices, be rude to jews, to red-heads, or to taxi-drivers. 

A single inconsistency in behavior may mean very little. A vigo- 
rous executive who at the office, at home, or wherever he is in com- 
mand, makes swift and emphatic decisions, may be reduced to virtual 
immobility when confronted in a restaurant with a tray of French 
pastry. Why? Perhaps it is just fatigue at the end of the day, or 
annoyance at being forced to decide so trivial an issue (in such cases 
the trait of decisiveness is not aroused) ; or his hesitation may be a 
throw back to some boyhood experience when the rod had inhibited 
the stealing of tarts. Such specific inhibitions may be at work, or the 
customary decisiveness may perhaps for this occasion be blocked by 
the activity of other deep-seated traits now functionally dominant, 
such as frugality or hypochondria. 

Or, take the case of Dr. D, always neat about his person and 
desk, punctilious about lecture notes, outlines, and files; his personal 
possessions are not only in order but carefully kept under lock and 
key. Dr. D is also in charge of the departmental library: in this duty 
he is careless; he leaves the library door unlocked, and books are lost; 
it does not bother him that dust accumulates. Does this contradiction 
in behavior mean that D lacks traits? Not at all. He has two opposed 
stylistic traits, one of orderliness and one of disorderliness. Pursuing 
the case further, this duality is explained by the fact that D has one 
cardinal (motivational) trait from which these contrasting styles 
proceed. The outstanding fact about his personality is that he is a 
self-centered egotist who never acts for other people’s interests, but 
always for his own. This cardinal trait of self-centeredness (for which 
there is abundant evidence) demands orderliness for himself, but not 
for others. 

For various reasons, therefore, simple statistical correspondence 
of measures is not the sum-total of the available evidence for traits. 
The more profound congruences of behavior emerge only after an 
intensive study of the organization of each personality. 

And yet in arguing that the inner consistency of traits is often 
greater than surface explorations indicate, we must not commit the 
opposite error of presuming every trait to be organically self-con- 
sistent under all circumstances. As has been pointed out, traits are 
often aroused in one type of situation and not in another; not all 


stimuli are equivalent in effectiveness. Successful adaptation and 
mastery require a trait to remain loose-knit, so that its determinative 
influence may be modified or checked according to special demands 
of the moment. To serve as a successful modus vivendi a trait must 
remain plastic or lose its usefulness. 

We conclude then that the consistency of a trait is entirely a 
matter of degree. 18 There must be some demonstrable relationship 
between separate acts before a trait can be inferred. Yet occurrence 
of dissociated, specific, or even contradictory acts is not necessarily 
fatal to the inference. Perfect and rigid self-consistency is not to be 


The question of the distribution of traits arises, of course, only 
in connection with common traits. By their very nature individual 
traits cannot be scaled in a population. 

Now, the investigator ‘wants to think of any common trait as a 
single continuous variable with scores arranged according to a normal 
curve of distribution, corresponding to the probability (Gaussian) 
curve, so that he may treat the scores for any common traits accord- 
ing to convenient statistical principles. To have such a distribution 
facilitates the establishment of norms, the comparison of one indi- 
vidual with another, computations of correlations with other vari- 
ables, and many other quantitative procedures dear to the heart of 
the tester. Without doubt, the most serviceable common traits are 
those whose distribution corresponds fairly closely to this require- 
ment. Figure 22 illustrates the situation with one scale, the continuum 
in this case being Ascendance-Submission, constructed from a pair 
of traits, initially conceived according to the logic of biological 
adaptation and acculturation (cf. pp. 298, 410 If.). Items designed to 
measure people on this continuum were scored in such a way that 

18 The reader should note how frequently it has been necessary to speak of 
various attributive characteristics of traits as 'a “question of degree.” Earlier, the 
amount of driving or of directive stress in a trait was shown to vary from case to 
case (as well as from time to time) ; further, a common trait was seen to be com- 
mon only to a certain extent and in a certain sense; then too, independence turned 
out to be only a relative attribute of traits; and now the self-consistency of a trait 
also seems to be a variable matter. Likewise, we shall soon see that the importance 
of a trait in an individual life may vary from a cardinal position to a minor and 
inconsequential one. Finally, the unity of personality itself, as the next chapter will 
show, is neither perfect nor wholly mythical; it too is a question of amount. No 
sharper distinctions or more final statements are possible, for the simple reason that 
mental organization in all its phases is a matter of degree! 


they yielded a fairly normal curve of distribution, the extreme scores 
at either terminus of the linear scale signifying opposite modes of 
adjustment. Statistically considered, there is a single variable; psycho- 
logically, however, the composition of this variable is by no means 

In defense of the view that these linear continua represent a single 
psychological variable, it is customary to cite the supposed prefer- 
ence of nature for symmetrical distributions. So many minute genetic 

Distribution of Scores from a Test Designed to Measure 

Constructed from the decile distribution of scores published in Manual of Direc- 
tions , rev. ed., for “A-S Reaction Study,” Form for Men. 

influences determine the height of a man, for example, that the final 
stature attained by a large unselected male population will be found 
to vary about a central tendency as chance itself varies. When multi- 
ple causes are at work the normal curve of variation frequently found 
in biological measures is, therefore, the curve of probability. And 
this is why the native ingredients in personality, aspects of physique, 
temperament, and nervous plasticity, dependent as they are upon the 
composite determinants of inheritance, frequently yield symmetrical 
curves of distribution. The more any quality depends upon chance- 
biological determination the more likely it is to be normally dis- 

Traits, however, do not depend altogether on chance-biological 
determination. They depend also upon cultural determination, and 
here an entirely opposed principle is at work, for, roughly speaking, 
in so far as a quality depends on the mores, the less likely it is to be 
normally distributed. Folkways, civil laws, and all other exterior con- 
straints tend to destroy the “natural” variation in behavior through 
their demand for conformity . 

For example, motorists approaching an intersection without external 
control, will vary their behavior according to chance factors. The 



curve on a continuum from “Same Speed Ahead— Full Stop” is a 
normal curve. But only when strong social pressure is lacking does 
such a Gaussian distribution appear. Let an element of social constraint 
enter— “Stop” signs, red lights, a policeman, and the distribution of 


Figure 23 

J-Curve of Conforming Behavior 

Motorists’ behavior at intersection, with no cross traffic approaching, but con- 
fronted with red lights and a traffic officer: 102 cases. After F. H. Allport, /. Soc. 
Psychol., 1934, j, p. 144, Figure 2. 

speeds is no longer “normal.” It is skewed markedly toward the side 
of obedience. A large number of instances of social conformity have 
been studied, always with the same result. 18 What appears in place of 
a normal (chance-biological) curve is a J-shaped distribution. The 
behavior of motorists confronted physically by stop lights and a 
policeman is represented in Figure 23. 

19 F. H. Allport, “The J-Curve Hypothesis of Conforming Behavior,” J. Soc. 
Psychol., 1934, 5. 141-183. 


In many situations, a double J-curve is found. Some strong social 
pressure makes for conformity; ever}' individual is expected to fit the 
same mold; and most of them do. But at the same time “uncontrolled” 
factors again enter to a sufficient extent to cause minor deviations 
from the norm. A good instance is found in the punching of the time 
clock in a factory, Figure 24. Here every incentive leads to punctual- 
ity, or if any deviation is permitted, it must be in the direction of 
earliness. The resulting distribution is sharply asymmetrical, with the 
mode falling upon the “punching time” set by the employer. 




=n rr'N 


Figure 24 

Double J-Curve of Conforming Behavior 

Smoothed curve representing time of arrival at a factory: 1,277 cases. Aftei 
F. H. Allport, op. cit., p. 145, Figure 5A. 

Underlying biological variation, then, makes for a distribution 
approaching the normal probability curve; cultural conformity, also 
operating in the case of common traits, tends, on the other hand, to 
make the distribution asymmetrical, with the mode at the terminal 
step defined by the norm itself. How can these opposed forms of 
distribution be reconciled? 

Before attempting a solution of the puzzle, we must ask ourselves 
one other question: Is it more reasonable to suppose in the case of a 
given scalable variable that we are confronted with one common 
trait or with two? Taking Ascendance-Submission as typical, is one 
aspect of behavior here in question (as the hyphenated name for the 
continuum implies) or are two different aspects arbitrarily brought 
together for convenience of measurement? The second alternative is 
the more probable. For submission is not merely the absence of 
ascendance; a high score on the former does not mean simply an 
absence of the latter. Submission is a positive manner of adjusting, at 



positive as ascendance itself. Many people have a well integrated dis- 
position to yield, to take the passive role in face-to-face situations; 
it is their style of life. The conception of a common trait of sub 
mission does its best to approximate this tendency as it is found in 
individual lives. And the same may be said of most other commonly 
scaled traits. Introversion is as positive (in the psychological sense) 
as extroversion; radicalism is as definite a common trait as conserva- 
tism. In such linear continua clearly two traits are involved. 20 Putting 
the two together is justified, not because they are one and the same 
trait, but because, from the point of view of adaptive significance, 
they are complementary . 

In the light of these considerations, the following suggestions are 
offered to account for the quasi-normal distribution of continua of 
the type represented in Figure 22. Two common traits are present, 
but since they represent aspects of adjustment that may be regarded 
as complementary, they are joined into one linear continuum. The 
true distribution of either trait considered alone is not normal. It 
approaches the J-curve of cultural behavior: most people do not de- 
part markedly from the norm that is culturally established; social 
pressure in these cases tends to produce moderation. Yet owing to 
special sets of influences (of which biological temperament is surely 
one) some individuals do deviate from the modal tendency. 

In the case of Ascendance-Submission our hypothesis is that two 
separate distributions are represented in Figure 22. Most people are 
neither decidedly ascendant nor decidedly submissive; their average 
or near-average score on the test signifies as much. 21 When the two 

20 It is not true, however, that common traits, as conceived by testers, are al- 
ways a combination of two opposite modes of adjustment. There are different ways 
of conceiving common traits. In some scales mere presence or absence of a tend- 
ency is measured, as, for example, in the case of esthetic, political, or religious 
interests , or in cases where the amounts of some prejudice or ambition are deter- 
mined. In these instances a high score signifies the presence of the trait in question; 
a low score merely its absence. The present argument applies to scales, for either 
traits or attitudes, that are represented by hyphenated names, or could be so repre- 
sented, radicalism-conservatism, expansiveness-reticence, introversion-extrover- 
sion, fascism-communism, militarism-pacifism, and the like. 

21 An analysis of the reasons for the large proportion of average scores on the 
A-S Reaction Study reveals many contributing factors. In the first place the primary 
influence seems to be a tendency for the subject to select moderate responses to 
virtually every question; that is to say, he shows himself to be consistently “average” 
(neither ascendant nor submissive) in most situations. Here in all probability is a 
double effect: on the one hand, the social norms favor people neither markedly 
ascendant nor markedly submissive; on the other, the chance factors of tempera- 
ment and other biological causes tend by and large in the same direction. A sec- 
ond, less frequent, cause of average scores is the cancellation of strongly ascencbnt 


traits are pat into a single artificial continuum, the result is a quasi- 
normal curve. If the distribution is still asymmetrical, as it often is, 
the inventor of the scale alters one item and another until a more 
satisfactory curve is obtained. The last step-so completely arbitrary- 
is often forgotten by the inventor when he interprets his results. He 
overlooks the fact that he personally has had much to say about 
the distribution which he blithely attributes to nature. 

By way of summary: The whole problem of the distribution of 
common traits is exceedingly complex. Though for the present it is 
well to leave it open, a few suggestions have been made concerning 
factors influencing the distribution. For one thing the “normal” curve 
for common traits does in part reflect nature’s preference for average 
(non-distinctive) levels; but in addition it reflects social pressure 
toward conformity with some accepted “average” level of conduct. 
The normal distribution often results from an arbitrary juxtaposition 
of opposite modes of adjustment into one linear scale. Finally, the 
distribution is to a certain extent affected by the inventor of the test 
who sees to it that successive revisions of his scale yield a more and 
more symmetrical scatter of scores. There is no serious objection to 
this last procedure, nor to the normal curve thus obtained, since in 
any case the scaling of common traits is but a rough way of approach- 
ing personality. In the interests of sound theory, however, it is well 
to realize the extent of the complications introduced into one “simple” 
variable for purposes of convenient measurement. Only by so doing 
can we avoid the fallacy of presuming, as more than one writer has 
done, that “everything in personality is normally distributed.” 


In every personality there are traits of major significance and traits 
of minor significance. Occasionally some trait is so pervasive and so 
outstanding in a life that it deserves to be called the cardinal trait. 
It is so dominant that there are few activities that cannot be traced 

responses by an equal number of markedly submissive responses. In such cases, the 
scale is misapplied and misleading, for the subject obviously is not “average” in any 
intelligible sense of the word. He has two well integrated but entirely contradictory 
traits. Finally, an average score might result (though it seldom does so) from an 
entirely random series of responses, whereby this ascendant reply is cancelled by 
that submissive reply, with a number of non-diagnostic (moderate) responses 
thrown in, all as the theory of specificity would require. Thus there are many causes 
of average scores, and it is for this reason that an average score obtained by * 
single individual has little psychological significance. 


directly or indirectly to its influence. The list of terms on pp. 302 f., 
derived from the proper names of historical and fictional characters, 
shows clearly what is meant by cardinal traits. No such trait can for 
long remain hidden; an individual is known by it, and may even 
become famous for it. Such a master quality has sometimes been 
called the eminent trait , the ruling passion, the mister-sentiment, or 
the radix of a life. 

It has been objected that the conception of a cardinal trait is 
essentially tautological, for, one asks, is not the cardinal trait identical 
with the personality itself? This objection cannot be admitted; how- 
ever well integrated a life may be around the cardinal trait there re- 
main specific habits, incidental and non-organized tendencies, and 
minor traits of some degree that cannot be subsumed functionally 
under the cardinal trait. Though pervasive and pivotal, a cardinal 
trait still remains within the personality; it never coincides with it. 

It is an unusual personality that possesses one and only one emi- 
nent trait. Ordinarily it seems that the foci of personality (though 
not wholly separate from one another) lie in a handful of distin- 
guishable central traits. (Cf. Perry’s list of the four “benign” traits 
of William James, p. 314, ftn. 2.) Central traits are those usually 
mentioned in careful letters of recommendation, in rating scales 
where the rater stars the outstanding characteristics of the individual, 
or in brief verbal descriptions of a person. 

One may speak, on a still less important level, of secondary traits, 
less conspicuous, less generalized, less consistent, and less often called 
into play than central traits. They are aroused by a narrower range 
of equivalent stimuli and they issue into a narrower range of equiva- 
lent responses. Being so circumscribed they may escape the notice of 
all but close acquaintances. 

It goes without saying that these three gradations are altogether 
arbitrary and are phrased merely for the convenience of discourse. 
There are no criteria, statistical or otherwise, by which to mark off 
one grade from another. In reality there are all possible degrees of 
organization in a trait from the most circumscribed and unstable to 
the most pervasive and most firmly integrated. It is useful, however, 
to have these distinctions at hand when one wishes to speak roughly 
of the relative prominence of various traits in a given personality. 




Since semi-separate traits are the most reasonable units for use in 
the psychological exploration of personality, it has been necessary to 
discuss them in some detail. It would, however, be unfortunate to 
give the impression that a personality is adequately studied when a 
list of these inter-dependent traits is assembled. The problem of sub- 
structures is not the same as the problem of total structure. Traits 
are merely the principal Teilfunktionen of personality; beyond them 
extends the problem of their inter-fusion. Various theories have been 
advanced to account for the homogeneity of the whole personality 
which Stern has so aptly characterized as an imitas multiplex. Thus 
far we have been considering theories that explain its multiplicity; 
the next question in order is its unity. Be it noted, however, that the 
theory of inter-dependent traits takes a long step in the direction of 


In everyday life, no one, not even a psychologist, doubts that 
underlying the conduct of a mature person there are characteristic 
dispositions or traits. His enthusiasms, interests, and styles of expres- 
sion are far too self-consistent and plainly patterned to be accounted 
for in terms of specific habits or identical elements. Nor can the 
stability and consistency of behavior be explained away by invoking 
nominalistic theories; stability and consistency are not due to the bio- 
social arrangement of unrelated activities into categories with verbal 
tags. Traits are not creations in the mind of the observer, nor are 
they verbal fictions; they are here accepted as biophysical facts, 
actual psychophysical dispositions related— though no one yet knows 
how— to persistent neural systems of stress and determination. 

Traits are not, like the faculties of old, abstractions derived from 
a theory of mind-in-general. There is no essential resemblance be- 
tween impersonal faculties, as Memory, Will, and Sagacity on the one 
hand, and the focalized sub-structures of a particular mind (interests, 
sentiments, general attitudes) on the other. Faculties are universal, 
traits personal; faculties are independent, traits inter-dependent; facul- 
ties are a priori, traits must be ascertained empirically in the indi- 
vidual case. 

The doctrine of traits differs also from the theory of factors or 



any other system of common dimensions into which every indi- 
vidual is fitted categorically. Conceptualized nomothetic units 
(factors, instincts, needs, and the like) stress what is universal in men, 
not what is organized into integral, personal systems. The doctrine 
of traits emphasizes concrete individuality. 

Traits are not directly observable; they are inferred (as any kind 
of determining tendency is inferred). Without such an inference the 
stability and consistency of personal behavior could not possibly be 
explained. Any specific action is a product of innumerable determi- 
nants, not only of traits but of momentary pressures and specialized 
influences. But it is the repeated occurrence of actions having the 
same significance (equivalence of response), following upon a defin- 
able range of stimuli having the same personal significance (equiva- 
lence of stimuli), that makes necessary the postulation of traits as 
states of Being. Traits are not at all times active, but they are per- 
sistent even when latent, and are distinguished by low thresholds of 

It is one thing to admit traits as the most acceptable unit for in- 
vestigation in the psychology of personality, but another to deter- 
mine authoritatively the precise character of these traits in a given 
life. In order to avoid projection of his own nature and many other 
sources of error, the psychological investigator must use all the em- 
pirical tools of his science to make his inferences valid. Traits can- 
not be conjured into existence; they must be discovered. 

In naming the traits that are discovered, there are many pitfalls, 
the chief one being the confusion of personality with character 
through the use of eulogistic and dyslogistic terms. Whenever this 
occurs the existential pattern of personality becomes hopelessly en- 
tangled with social judgments of merit and demerit. It is possible, 
though difficult, to achieve a psychological vocabulary of non- 
censorial trait-names. Most of these terms antedate psychology by 
centuries; they were invented because they were needed. Simply be- 
cause it is difficult to employ them circumspectly, the investigator is 
not on that account justified in dispensing with them altogether and 
attempting to put mathematical or artificial symbols in their place. 
Regrettable though it may seem, the attributes of human personality 
can be depicted only with the aid of common speech, for it alone pos- 
sesses the requisite flexibility, subtlety, and established intelligibility. 

For the purposes of comparison and measurement certain seg- 
ments of behavior (by virtue of the similarity of human equipment 

RESUME 34 i 

and the common exigencies of the cultural and physical environ- 
ments) may be considered as distributed in a general population. 
These common traits as conceptualized by the investigator may, in 
a rough and approximate way, be scaled on a linear continuum. The 
“normality” of distribution obtained for such traits is a complex prod- 
uct of chance-biological variation, cultural conformity, and artifact. 
However carefully conceived and scaled, a common trait is at best 
an abstraction, for in its concrete form, in each particular life, it 
operates always in a unique fashion. Individual traits cannot be scaled 
at all in a general population, and for that reason they have been 
hitherto neglected by all excepting clinical investigators. 

Some traits are clearly motivational, especially those sub-classes 
ordinarily known as interests, ambitions, complexes, and sentiments. 
Other traits are less dynamic in their operation, having an ability to 
steer (to stylize) behavior rather than to initiate it. But often the 
traits that are at first directive acquire driving power, and those that 
are at one time driving become merely directive. 

Traits are not wholly independent of one another; nor are any 
other neuropsychic systems. They frequently exist in clusters, the 
arousal of one portion tending to spread to all regions in readiness 
for communication. Throughout this elaborate interplay, different 
foci of organization can be detected, a fact that justifies the concep- 
tion of a manifold of traits even where they clearly overlap. 

As this segregation among traits is only relative, so too is their 
self-consistency. In fact, the usefulness of any trait to its possessor 
depends to a great extent upon its flexibility. Even while it stabilizes 
conduct and economizes effort, the trait must not be rigid in its opera- 
tion; for effective adjustment and mastery require variation. The 
range of situations that arouses traits must be expected to change 
according to circumstances. Also in any personality one must expect 
to find some contradiction and conflict among traits, as novelists and 
clinicians never tire of telling. 

Variable though they are, still in every mature personality certain 
central traits can normally be identified. So too can secondary traits, 
though these are less distinctive, less prominent, and more circum- 
scribed in their operation. Whenever a disposition is so little general- 
ized that it is aroused by only a narrow range of stimulus situations, 
it is more properly called an attitude than a trait. Somewhat rarely 
a personality is dominated by one outstanding cardinal trait, to which 
other dispositions serve as merely subsidiary, congruent foci. 



As is the case with all other forms of mental organization, the 
structure of true (individual) traits is a question of degree. But how- 
ever much they may vary in respect to their consistency, scope, and 
independence, they have— according to the theory developed in these 
chapters— certain essential characteristics. They are always biophysi- 
cal in nature, concrete and personal in their organization, contempo- 
raneous in their effect, capable of functional autonomy, but not struc- 
turally independent of one another; they are generalized (to the ex- 
tent that the effective stimuli are equivalent, and to the extent that 
the resultant responses are equivalent). They are modi vivendi, ulti- 
mately deriving their significance from the role they play in advanc- 
ing adaptation within, and mastery of, the personal environment. 



Made up of myriads of microscopic cell lives, individually 
born, each one of us nevertheless appears to himself a single 
entity, a unity experiencing and acting as one individual. In a 
way, the more far-reaching and many-sided the reactions of 
which a mind is capable, the more need, as well as the more 
scope, for their consolidation into one. True, each one of us is 
in some sense not one self, but a multiple system of selves. Yet 
how closely those selves are united and integrated to one per- 

— C. S. Sherrington 

In spite of the common preference of the psychologist for the 
analytical approach to any problem, he has not altogether neglected 
the task of accounting as well as he may for the idiomatic complete- 
ness of personality. Indeed, he has examined the problem from a great 
many angles, and has devised many theories and hypotheses, some 
of which he has submitted to empirical tests. Since the problem thus 
turns out to be many-sided, our treatment of it will of necessity be 
compendious and eclectic. 

The only approach that may safely be excluded is the rhapsodic. 
Here we find theories that do little more than assert personality to be 
an “Indivisible Whole,” “a total integrated pattern of behavior,” an 
Unteilbarkeit, an in sich geschlossme Ganzheit. This rapturous litera- 
ture of wholeness does not explore the unity that it apotheosizes; it 
merely contemplates and admires. Personality, it says, is like a sym- 
phony. Granted; but does not the comprehension of symphonic unity 
come only through an understanding of the articulate 'weaving of 
motifs, movements, bridge-passages, modulations, contrasts, and 
codas? Nothing but empty and vague adjectives can be used to char- 
acterize the work as a whole. If a totality is not articulated, it is 
likely to be an incomprehensible blur; it can then be extolled, but not 
understood. What is more fatal, the rhapsodic approach seriously 



over-simplifies the whole problem, under-estimating the conflicts and 
discords in ever} 7 life. Unity, at best, is a matter of degree . 1 


The problem of unity, as we have said, is many-sided. Here and 
there in the last nine chapters it has been incidentally discussed. 
Before considering it from new points of view, a backward glance 
will recall the issues as they have thus far arisen. 

In infancy there is marked evidence of a “dynamical unity” that 
seems never in later life to be of quite such striking degree. The 
young child responds pretty much “as a whole” whenever he reaches, 
retracts, or expresses emotion. The situation was represented aftei 
the manner of Lewin in Figure 14 (p. 133) where it appeared that 
few psychical systems are segregated in the earliest months, and that 
the barriers between these are weak. Even after the first months of 
life the child responds for a long while in what seems almost an “all 
or none” fashion, particularly when emotion is involved. He is in- 
capable of delayed response, of discrimination, and of precise grada- 
tion. As he grows older and as differentiation within the psychical 
systems takes place this elementary dynamical unity is lessened, 
though vestiges remain throughout life in states of emotional seizure, 
and in the persistence of synkinesis (patterns of expressive movement 
undifferentiated from one another) . 

As if to offset the disunity that comes with differentiation, there 
is a compensatory process of integration. By virtue of the functional 
joining of psychical systems (through conditioning, generalization of 
habit, and ail associational processes) integral units come into ex- 
istence. For the most part these units represent coherent foci of de- 
velopment, found serviceable to adjustment and to mastery. Figure 1 5 
(p. 1 41) represents this multiplicity of units in an hierarchical ar- 
rangement, and shows that these functional units though to some 
extent independent, tend, nevertheless, normally to converge into 
more embracing systems. Though perfect unity is never achieved, 
there may be said to be a constant progression in that direction, 

1 “The greatest difficulty in the scientific study of character comes with our 
realization that men by no means always behave as if their mental life bore a single 
stamp. Each performs acts or speaks words that are ‘uncharacteristic/ seeming in- 
deed to run counter to the character we ascribe to him. No man is a machine 
whose functioning is perfectly predictable and whose performance one may order 
to a mathematical formula. But despite all this, neither is man a wholly chaotic 
creation.” R. Miiller-Freienfels, Lebemnahe Charakterkunde, 1935, p. 13. 



hindered however by the contrary formation and fixation of stereo- 
types, dissociated complexes, and primitive levels of adaptation. 

While differentiation and integration are under way there de- 
velops gradually an important core of self-consciousness. Perhaps 
nothing contributes to the unity of personality as much as this sub- 
jective point of reference, by virtue of which the individual feels 
that there is coherence between his memories of the past and his plans 
for the future. Self-consciousness is necessary for self-esteem, for 
aspiration, and is a pre-condition of status in the social group (by 
which, in turn, it is profoundly modified). All these factors are unify- 
ing in their effects. Gradually the self extends in such a way that it is 
closely identified with personal possessions, with other people, and 
with introcepted ideals and cultural standards. The self becomes the 
center of an orderly psychological universe. Whether the self is 
regarded as the innermost nucleus of all conscious ego-systems 
(Koffka) or as the interplay of all conscious states (James), does not 
greatly matter. In either case the self is the subjective moderator of 
whatever unity the personality may have. 

Many writers have stressed particularly the role of memory, 
pointing to its bridge-like character, reaching backward to provide 
easy communication between by-gone experiences. These bridges of 
memory can be completely broken only by an hysterical splitting of 
personality. And pathological dissociation of this type, as Sherring- 
ton says, bears important testimony to the inescapable unity of 
normal memory schemata: 

Even in those extremes of so-called double personality, one of their 
mystifying features is that the individual seems to himself at any one 
time wholly either this personality or that, never the two commingled. 
The view that regards hysteria as mental dissociation illustrates the 
integrative trend of the total healthy mind . 2 

In order to bind the past with the future, memory must be supple- 
mented by imagination , another unifying capacity of the self. With 
its aid the human being may plan his life when he is young, and 
spend years of concerted effort in pursuit of his chosen goals. 

Perhaps the most significant property of the self is the peculiar 
inward quality of emotional life, represented variously as the princi- 
ple of Egoism, self-esteem, the sentiment of self-regard, or as the “up- 

2 C. S. Sherrington, Mental Hygiene , 1923, 7, p, 16. This quotation follows im« 
mediately the passage cited at the beginning of this chapter. 



ward tendency of the Ego.” Whenever the beloved ego is the object 
of regard, as it very often is, unity is enhanced, for at such moments 
all activities have a clear common point of reference. 

Such selfishness, however, is not the whole story. An individual 
who devotes himself to one master-sentiment, whose personality is 
distinguished by one primary Bestmmmng (Biihler) likewise finds 
psychological unity. In fact the pursuit of external goals can be more 
consistently maintained than the opportunistic pursuit of selfish ends 
which of necessity vary from time to time. This conception of inten- 
tion as a principle of unification is related also to the conception of 
the ego-ideal (pp. 218-220). Whenever the ego-ideal is derived by 
virtue of introception from the ethics of culture, it helps to hold the 
individual within a single course of development. Any Weltan- 
schauung, however derived, by engendering intelligibility upon the 
diversity of experience, serves as an important unifying influence. 

Then there are the numerous balancing agencies in the body that 
preserve functional integrity in the course of growth: the homeo- 
stasis of the endocrines, the recovery or transfer of functions after 
injury, the remarkable adaptive properties of the sense-organs. All 
these biological conditions of unity must not be overlooked, for 
nature appears greatly concerned in preserving the integrity of the 
individual organism. This fact is the most basic of all the guarantees 
of unity. Hereditary endowment in terms of temperament contributes 
to this stabilizing of the course of development, with the result that 
as a personality changes it seems to change consistently with itself 
(cf. Figure 13, pp. 126 f.). Nature sets limits beyond which the varia- 
tion in individual development may not extend. 

In an entirely different sense, unity (of the moment) is guaran- 
teed by the principle of convergence. At any one time the available 
psychophysical energies are mobilized in one maximally integrated 
course of conduct, through the medium of the final common path 
(p. 313). Important as this type of unification is, it has two limita- 
tions. In the first place, it does not mean, as is sometimes mistakenly 
held, that the “organism responds as a whole.” Such a statement is 
an exaggeration (unless perhaps in the earliest days of infancy where 
mass action dominates the scene of behavior). Many concurrent 
activities can go along at one and the same time. A man may walk, 
smoke, dodge the traffic, digest his dinner, and at the same time be 
busy with his thoughts. The final common path means only that one 
maximally integrated activity occurs at a time. The man cannot be 



pursuing several different trains of thought at one and the same 
moment. The second limitation to the principle of convergence is 
that at best it guarantees unity only for the moment. In itself it does 
not account for the consistency between successive acts. To make 
good this lack, there must be an account of structural consistency 
whereby unity over a period of time is achieved. 

The hypothesis of traits fills this need. Far better than any other 
doctrine of elements this hypothesis helps to account for consistency 
in personality wherever consistency is found, and for inconsistency 
whenever conflict and discordance prevail. There is stability among 
traits; there is likewise contradiction; there are cardinal and central 
traits to which minor traits may be ordered as subsidiary foci, round- 
ing out or specializing the operation of the principal traits; and there 
are dissociated traits. Further, the conception of equivalence supplies 
the elasticity necessary to account for the variability of behavior 
within a range. Unity is, after all, never mechanically rigid; it shows 
itself only within a range of equivalent conduct. 

Most significant of all the attributes of traits for the problem of 
unity is their interdependence. Any theory of separate and inde- 
pendent elements will find itself hopelessly confused in attempting 
to represent the coherence among these elements (cf. Figure 17a, 
p. 246). Factorial psychology, for example, is faced with two alterna- 
tives, neither palatable. Either it may assume that personality as a 
whole is simply the sum-total of the independent factors, thereby 
committing the fallacy of the omnibus (bundle) conception of per- 
sonality; or else it must find some cement to make the separate and 
unrelated factors cohere. With the assumption of over-lapping traits 
the task is easier, for the very conception itself contains the idea of 
fusion. Unity lies in the overlap of the traits with one another. 3 This 

3 A. Kronfeld has considered the various ways in which the Teilfunktionen , or 
traits, might be related to the total personality. The first possibility is the formal 
summation (omnibus theory) that he rightly rejects. The second is the view that 
each trait may contain, microcosmically, the essential qualities of the whole per- 
sonality. This view seems to confuse convergence in action (the final common 
path) which is sound, with a curious and untenable principle of immanence, whereby 
the whole is regarded as resident in each part. It seems wiser to regard the disposi- 
tions as incomplete portions of the total personality, and to admit their convergence 
merely in action, where indeed one little act is often diagnostic of many qualities— 
as when we say, “That little gesture of his spoke volumes.” The third possibility, 
according to Kronfeld, is the one here advocated, namely, that the unity of per- 
sonality consists in the intricate functional inter-relationship of the traits, and thek 
blended arrangement in hierarchies. A. Kronfeld, Lehrbuch der Charakterkunde s 
I 93 2 ? p- 35* 



statement does not, of course, settle the issue for all time, since the 
nature of the overlap and its raison d'etre need to be known. But 
intrinsically any hypothesis of interdependent traits is far more ad- 
vantageous in the study of unity than an hypothesis of disparate and 
scattered elements. 

All these considerations have been fully discussed in earlier chap- 
ters. They are important as factors making for coherence in person- 
ality. But they do not tell the whole story. 

It is the three-fold argument of this chapter that the problem of 
unity must be approached from many sides; that the most significant 
contributions come with the aid of concepts from a somewhat more 
complex level than psychologists are wont to use; and, finally, that 
an empirical solution to many of the issues involved is possible. But 
before running the ever-present danger on the empirical road of 
losing one’s way in a welter of detail, it will be instructive to hear 
what some of the rationalistic writers say concerning unity. Hypothe- 
ses originating with rationalists are often the best guides for psycholo- 
gists when they enter new or only partially explored territory. 


As Chapter II made clear, there are many meanings of the term 
personality other than the special psychological meaning adopted in 
this volume. Metaphysically one might define personality as “the in- 
destructible essence of individual being (the soul).” Such a definition 
would of necessity claim unity as an essential attribute of personality. 
This proposition, though it may very well be true, lies altogether in a 
nonpsychological realm of discourse. In deciding to dispense with 
the soul, psychologists cut themselves off deliberately from such 
speculative propositions of theology and philosophy. Unfortunately 
they fell at the same time into the rather shabby habit of declaring 
that all the metaphysical conceptions with which they refused to 
deal were ipso facto meaningless. It would be far wiser to concede 
that metaphysical unity may be a property of personality, while in- 
sisting that it is a different problem from the empirical unity that 
falls in the province of psychology. 

Somewhat closer to the psychologist’s interest are those philoso- 
phies of nature variously called organismic, systemic, or holistic. 
Such philosophies see nature as composed of coherent systems fraught 
with significance and survival value; they frequently cite personality 



as the supreme example. Such was the view of Fechner who regarded 
any organism as an expression of the primary law of the “tendency 
to stability.” According to this principle any single system assumes 
a regular internal arrangement of its parts, and a regular external 
form whose stability is guaranteed by the very mandate of Nature . 4 
Spinoza too had a philosophy of orderly systems, the key to which 
lay in his conception of conatus. In the material -world an object at 
rest or in motion expresses its conatus se conservtmdi by resisting 
whatever would tend to change its state of motion or rest. In human 
beings the principle expresses itself rather as an active impulse toward 
growth and self-expression. Man is unique in that he alone is con- 
scious of the possession of this conatus* There are philosophies of 
system and order that regard the unification of personality as an 
illustrative event in a cosmos given to orderly and coherent arrange- 
ments; the philosophy of Whitehead being one of these. 

Even physiologists often embrace a philosophy of “systematic 
relevance.” The self-preserving, self-repairing, self-regulating proc- 
esses of the body imply to them a root-tendency to maintain whole- 
ness. In the constant return of all psychophysical systems to a state 
of equilibrium, some see a “wisdom of the body,” others a “state of 
vigilance.” The more prosaic refer merely to “homeostasis.” But 
whatever terms they employ these physiological doctrines all assume 
an inherent tendency of every organism to form itself into one intri- 
cate homogeneous system. 

Some philosophers, notably Keyserling, have commented upon 
the great difficulty, when compared with lower animals, that human 
beings have in achieving unity. Unity, they say, is attained only when 
all the possibilities of life have been realized. Lower animals are a 
comprehensive expression of their potentialities; human beings are 
seldom so. A kingfisher alertly catching food on the shores of a pond 
is a more perfect, if less intricate, unity than is a human being in 
quest of his daily bread. 


Romantics of the nineteenth century were fond of saying, “A 
man is what he loves.” Sometimes the adage was varied: “Everyman 

4 G. Fechner, Einige Ideen zur Schopfungs- uni Entwicklungsgeschichte dei 
Organismen , 1873, chap. iii. 

5 B. Spinoza, Ethics , Bk. II, chan. iii. sec. 2. 


bears the stamp of his favorite experience.” In recent years there 
have appeared somewhat more technical versions of the same teach- 
ing. Psychiatrists, for example, tell us that the normal personality is 
one that is ever active in pursuit of chosen goals; whereas abnormality 
is characterized by apathy, by a deficiency in life interests. Without 
aspirations a life cannot be steered in a consistent course. In many 
of the neuroses and psychoses a pathological lethargy settles upon 
the patient, and a deadening of personality results. 

Stating the case still more specifically, some writers find the unity 
of personality achieved in its life work. Burnham has pointed out 
that just as a concrete task (the Aufgabe ) integrates available energy 
for the moment, so too, in the long run, a life work confers stability 
and consistency. Voltaire gave classic expression to the thought when 
he maintained that salvation for the individual lies in “cultivating his 
garden.” The same view is upheld by John Dewey, who, however, 
hastens to add that there should be no fence about the garden, that 
the scope of one’s devotion must be ever widening. Goethe, too, found 
eventual salvation for Faust in useful work that finally absorbed 
Faust’s restless energies, and provided as nearly as possible the com- 
pleteness that he sought. 

In the course of striving it is natural that the “guiding image” 
or the ego-ideal should play a prominent part. A person centers 
his efforts on becoming what he wants to become, and develops con- 
sistently in ways that in his belief lead to that end. The guiding 
image fixes attention in one direction, dictates the skill that he must 
acquire, and prescribes the criteria by which all possible courses of 
conduct must be tested before they are engaged in. Even if the im- 
mediate details of a life are discordant, even if there seems to the out- 
sider to be a bewildering clash of purposes, still there may be an 
imaginal unity among them that gives subjective coherence in spite 
of all visible disharmony. 

Here is a curious fact: the attainment of unity depends more 
upon knowing what one wants than upon getting it. It is the striving 
towards the known goal that confers unity, not the successful arrival. 
Love of learning— to take an example— is more of a unifying force 
than the possession of learning, so too is love of art, of money, or of 
fame. Unity of intention offsets failure of accomplishment; it is a 
matter of what a man loves, not of what he has or acquires. Attain- 
ment may even be destructive of unity, for attainment forthwith 
abolishes the unifying desire. From this point of view unity lies only 


in the struggle for unity. It was Mephisto’s wager that he could so 
beguile Faust that the latter would no longer struggle for complete- 
ness, but surrender to some tempting state of self-satisfaction along 
the way. Had Faust yielded to the illusion that his quest was ended 
he would have been damned. In the end he was saved because he 
ceaselessly strove for the completeness he never attained. 

Wer imner strebend sich bermiht 
Den konnen wir erlosen. 

Faust, as the prototype of man, found that striving for complete- 
ness was not merely an abstract matter. The only practicable condi- 
tion of unity that he discovered was the seeldng of specific objectives 
related to a life work. When every moment of effort is directly or in- 
directly pointed to the same progressive series of goals, these moments 
are then bound to one another. Such an interlocking series of mo- 
ments constitutes what Paulhan has called a “harmony of striving,” 
and serves as the prime condition of unity in personality. The har- 
mony is rarely perfect, for in most lives aims clash with one another 
quite as readily as they reinforce one another. 6 From the psycho- 
logical point of view such a clash of purposes is inimical to unity. 7 


Many of the factors reviewed in the preceding pages can be sub- 
mitted to empirical study. This is fortunately true, for since unity is 
a matter of degree, it is of considerable importance to have methods 
for determining in a given case what degree has been attained. 

Empirical studies of unity, though still in the early stage of de- 

6 Lasurski rightly warns against mistaking pseudo-unity for true unity. The 
former is an illusory product of suppression and dissociation, “which consists in the 
fact that some inclination or group of related inclinations control all others, in- 
hibiting them or suppressing them. Men who are given to self-denial and asceticism, 
often serve as examples of this specious unity.” A. Lasurski, “Uber das Studium der 
Individualist,” Padag . Monog., 1912, No. 14, 27. Freudian psychology deals at length 
with the duplicity and lack of genuine unity in such personalities. 

7 Metaphysically, this statement may be denied. W. Stern has argued that since 
the clash of different aims takes place entirely within the person (he being in con- 
flict with himself), the person, therefore, must be an embracing unity, capable of 
harboring all conflicts that occur. Likewise, says Stern, the existence of personal 
conflict presupposes an ideal of unity for the person, or else a state of prior unity 
or a unity of the future; otherwise a conflict could not be evaluated as such; it 
would have no meaning. Allgemeine Psychologies 1935, p. 623. In other words, the 
person, ontologically considered, is more of a unity than he is from the empirical 
point of view. No doubt this is true, but it is only with the empirical (psychologi- 
cal) unity that we are here concerned. 


35 2 

velopment, have had the wholesome effect of exposing and correct- 
ing the exaggerations of unity ordinarily made in everyday life. As a 
rule we stereotype our judgments even of our most intimate friends, 
and see greater consistency in their behavior than we should. The 
reason for this over-simplification, of course, is the reason for all 
time-saving cliches: we cannot afford to think or to deal with ob- 
jects in all their intricate and conflicting aspects. Hence we seek the 
“essence,” and in so doing often arrive no further than a pigeon- 
hole . 8 It is for this reason that psychology with its corrective patience 
may improve upon the too simple perceptions of common sense. 

Empirically considered, the problem of unity is the same as the 
problem of consistency. The latter term is for experimental work 
much to be preferred, for it covers readily all approximations of unity, 
and can be applied to part-structures as well as to the whole. 

There follows a brief characterization of the principal methods 
now employed in psychology in the study of personal consistency. 

Prediction. Predictions of human conduct may be made under 
three conditions. First, when people are viewed en masse , and only 
the average behavior is of any interest. The experienced manager of a 
restaurant or moving picture theater can predict remarkably well 
how many people of the thousands who pass his establishment every 
hour will turn in at his door. The insurance company predicts ac- 
curately how many people will die or be injured in a given year. 
Such actuarial prediction, since it has nothing whatever to do with 
the individual, has no direct bearing upon the psychology of per- 
sonality. The second type of prediction, generally employed by the 
psychologist, comes only slightly closer. It is based upon knowledge 
of mind-in-general. The psychologist predicts that any man will 
blink his eye if the cornea is touched, or that any normal individual 
will show a gradual increase of proficiency while learning a motor 
skill. Prediction of this type is possible through the knowledge of 
the general properties of reflexes and habits; what is common in 
human nature affords the basis of the prediction. 

The third type of prediction, more relevant for the psychology 
of personality, forecasts what one individual man (and perhaps no 

8 In biographies, even those that are full length, an inevitable exaggeration of 
consistency occurs. “Irrelevant” activities and traits are discarded, and the act of 
discarding makes for over-simplification. Particularly clear is the exaggeration in 
the case of necrologies. The writer wishes to extract the “essence” or meaning of 
the life. In so doing remarkable unity emerges, more than was ever present in the 
animate person. 


one else) will do in a situation of a certain type. Such prophecies 
pertain to mind-in-particular, and are absolutely indispensable in 
ordinary life. It is only by virtue of them that we are able to select 
gifts that our friends will like, to bring together a congenial group at 
dinner, to choose words that will have the desired effect upon an 
acquaintance, or to pick a satisfactory employee, tenant, or room- 
mate. That our predictions sometimes go awry is true, but the fact 
that they are so often successful is one of the principal lines of evi- 
dence for the existence of relatively stable dispositions in personality. 

It is not, of course, the exact response of an individual that is 
predictable. It is only the range of his response. Rarely can we fore- 
tell the precise words our friend will use in expressing pleasure at the 
gift we have brought him, but that he will like it we are sure. With 
just what movements an aggressive person will show his nature in a 
given situation we do not know, but that he will be aggressive in 
some way we can safely wager. In other words we predict the opera- 
tion of a trait, but allow for a fairly wide range of equivalent re- 
sponses that will be called forth by other determinants prevailing at 
the time the behavior takes place. 

It is not difficult to bring this type of prediction into the labora- 
tory. One experiment based on this procedure was described on p. 
254. The experimenter predicted with considerable success which of 
his subjects would violate the prohibitions contained in the directions 
for their part in the experiment, though in all probability the investi- 
gator could not have foretold precisely when and how the violation 
would occur. It will be recalled that in this particular experiment, a 
rather special condition was introduced, namely, the making of the 
prediction on the evidence of first impressions (five minute ac- 
quaintance) . Even so, the predictions were correct in 69 per cent of 
the cases (chance = 50 per cent). Longer acquaintance, it may be 
assumed, would have enhanced the validity of the prediction. 

Another experiment, based on longer acquaintance, was conducted 
by Bender. 9 After four one-hour conferences with each of eight sub- 
jects, this investigator predicted the scores that each would make in 
ten different “personality tests.” When the tests were administered, 
and the 80 predictions correlated with the 80 performances, a co- 
efficient of +.55 ±.05 was obtained. 

After further acquaintance (sixteen one-hour conferences with 

9 1 . E. Bender, A Study in Integrations of Personalities by Prediction and Match- 
ing, 193 j (Syracuse University Library). 



each). Bender undertook with four new tests to predict not the total 
scores but the exact response each subject would make to every item. 
This feat, as pointed out above, is much more difficult than the pre- 
diction of total scores. Nevertheless the majority of the predicted 
responses were correct. 

A further variation in this work consisted in using 100 judges not 
personally acquainted with the subjects. These judges had for their 
guidance sketches of the eight personalities prepared by the experi- 
menter after his prolonged study of each case . 10 From each condensed 
portrait (called by the experimenter “integrational hypothesis”), the 
judges predicted what responses the eight subjects would make to a 
scale measuring various attitudes and interests (none of the attitudes 
or interests in question being mentioned specifically in the sketch). 
Even under this tenuous condition, success exceeded chance expecta- 
tion to a marked degree. 

These are only a few of the varieties of the many-sided method 
of prediction. It unquestionably demonstrates personal consistency in 
conduct, though it does not of course illuminate the question of the 
causes of consistency. As currently applied the method probably errs 
in the direction of under-stating the consistency of personality, rather 
than over-stating, for with the introduction of any judge into the 
research, his own failures to predict correctly may mistakenly be 
interpreted as a sign of inconsistency in the subject. Whenever judges 
are employed in investigations of personality it is found that they as 
human instruments vary greatly in their sensitiveness and skill (cf. 
Chapter XVIII). 

Matching. A related method, employing likewise the impression 
of the judge, has come into wide favor in recent years . 11 It is the 

10 One example follows and illustrates the type of “thumb-nail sketch” that 
psychologists sometimes offer as a rough synthesis of a personality they have stud^ 
ied by analytical methods. 

“K is characterized by a dignity of manner and bearing calculated to barricade 
her from certain contacts which she holds in disdain. She is interested in matrimony 
provided it will assure her economic security, luxury, travel and social prestige. Her 
reactions to people while often formal tend to develop into intensely personal con- 
tacts. With men, there is a high degree of sex-consciousness. She is selfish, i.e., self- 
centered, pleasure-seeking, dilettante in her interests rather than profound. For the 
greater part she is dissatisfied and bored with life. There is a tendency to be moody 
and sometimes languid. She is neat to the point of being fastidious. Although intro- 
verted she is ascendant. Her high social intelligence is well integrated around her 
striving for social approbation. What her personality lacks in force and seriousness 
of purpose is redeemed by her physical attraction and her personal charm.” 

11 Directions for using this method, as well as a valuable summary of research 
employing it, are given by P. E. Vernon, Psychol. Bull 1936, 33, 149-177. 



attempt to match different records of one personality by a judge 
who is confronted with an assortment of records taken from many 
personalities. The records may be of any type: life-histories, photo- 
graphs, specimens of handwriting, scores on various tests, artistic 
productions, or anything else. 

Bender, in the investigation cited above, used this method in addi- 
tion to prediction. In one experiment, 91 judges were given the eight 
thumb-nail sketches (“integrational hypotheses”) and a “percentile 
testograph” for each subject, constructed from the subject’s scores on 
a series of tests. The number of correct matchings obtained was com- 
pared with the number of correct matchings expected by chance; in 
six of the eight cases the judges showed that they clearly were able 
to order the two sets of records correctly; they perceived the unity 
that obtained between the summary sketch and the test scores. 

The method of matching has the advantage of permitting complex 
productions to be ordered with other complex productions, without 
first undergoing the destructive process of artificial analysis. To put 
the case another way, matching permits the quantitative study of 
qualitative patterns. Its utility will be further discussed in later chap- 

Correspondence of Measures. Both prediction and matching de- 
pend upon the ability of judges to order the records with which 
they deal. Some methods eliminate this source of error by compar- 
ing directly the record of an individual in one sphere of activity 
with his record in another. For example, through the use of correla- 
tion it is easy to determine whether the position of subjects relative 
to one another for one performance remains essentially the same for 
other performances. To take a typical instance: on p. 222 it was re- 
ported that for one group of subjects the correlation between meas- 
ures of insight and of humor was +.88. Interpreted, this coefficient 
means that for most individuals in the population studied, both quali- 
ties were present (or deficient) to about the same degree. (In this 
particular case the measures used were ratings, in which an initial 
error of subjective judgment, perhaps a “halo,” may be included, but 
some initial experimental error is possible, of course, even in objective 
measures.) Whenever the correlation between two or more qualities 
is high, consistency (barring initial errors) has been demonstrated 
between these qualities within the group as a whole; what is proved 
is that there is a tendency for the subjects to be consistent in the same 





%v ay. (The degree of consistency any single individual has achieved 
cannot, however, be determined by this method.) 

A somewhat similar form of correspondence was described on 
p. 316 where the internal consistency of any one scale for the 
measurement of a common trait was interpreted to mean that this 
trait (again within a total population) is self-consistent. This pro- 
cedure, too, tells nothing about the individual pattern of unity, but 
only that some common trait, as conceived by the experimenter, is to 
a certain degree self-consistent for a whole population of subjects. 
Likewise, the products of factor-analysis, determined by an extension 
of correlational methods, are empirically self-consistent. But they too 
pertain only to the abstract average man, and not to single indi- 

Through a more elaborate use of corresponding measures attempts 
have even been made to determine whether consistency (or its oppo- 
site, variability) is itself a consistent attribute of personality. The in- 
dividual’s variability of scores about his own mean score for many 
tests is taken as a kind of “integration index.” The conception is one 
of the “consistency of consistency” (or, its complement, the “con- 
sistency of variability”). Though offering a pretty problem for statis- 
ticians, this type of work has not as yet resulted in anything of 
psychological merit. 12 Since the unity of personality is not a matter 
of approximating the same “mean score” on many common traits, 
this procedure is maladapted to the problem in hand. 


There is a wide gap between the type of consistency demon- 
strated by the correspondence of measures, and the radical (root) 
consistency with which clinical psychology deals. The former looks 
at a large group of people and aims to discover what common 
clusters of coherent qualities there are for this population; the latter 
is solely interested in the congruence of an individual life. 

A young artist once took part in some psychological experiments 
in which many records were taken of his expressive movements, along 
with those of 24 other subjects. In seeking the correspondence of 
measures for all the variables studied, it turned out that the records 
of this particular subject were often “inconsistent.” For example, the 

12 Cf. G. W. Allport and P. E. Vernon, Studies in Expressive Movement, 1933, 
pp. 123-133. 


average pressure he exerted upon the point of a pencil in writing or 
upon the point of a stylus in tapping was 7th in the group; that is to 
say, the contact that he made upon a surface was somewhat firme r 
than the average. Yet records taken of the pressure he exerted in rest- 
ing his hand upon the surface, showed that his rank order in the group 
for this performance was 19th. Obviously there was no correspon- 
dence between these two measures: the point pressure was heavy; the 
pressure of resting the hand, light. No correlation was indicated". Was 
there no consistency? Certainly not in respect to pressure. But pres- 
sure is a superficial variable, a concept originating in physics and not 
in psychology. Further consideration suggests that though the meas- 
ures failed to correspond they are none the less congruent psycho- 
logically. Being an artist (and this is the unifying condition), this sub- 
ject, following his customary style of execution, rests his hand lightly 
upon his work, but plies his crayons or brush with some firmness upon 
the surface he is using. 13 

This simple illustration of congruence has many counterparts in 
the study of personality. The routine search for the correspondence 
of measures in order to establish some preconceived dimension of 
unity (in the above case, pressure ) is often disappointing. Sometimes 
it succeeds, but frequently does not. The weakness of this method is 
that it seeks consistency only in the horizontal dimension— between 
the peripheral activities of many people, and not in the vertical di- 
mension, down through the structure of the single personality (cf. 
p. 1 1 ) . By defining consistency only in terms of correspondence and 
never in terms of congruence a thoroughly one-sided picture (oi 
specificity) is likely to appear. 

A deeper consistency may often underlie superficially discordant 
activities. The case of D (p. 331) we recall as an instance in point. 
This man, a teacher, seemed one moment meticulous in his behavior, 
the next, careless and even slovenly. Measures of neatness in this case 
would certainly not correspond. But by looking further into the case, 
the illusion of specificity vanishes, for it appears that D is always 
orderly in respect to his personal possessions, and always disorderly 
in respect to other people’s. The first step in the analysis of congru- 
ence, then, establishes the existence of two opposite traits. Pursuing 
the case still further, even these opposites are reconciled through then 
relation to a single essential quality of self-centeredness peculiar to D, 
This root quality “explains” the inconsistency in his expressive traits. 

13 This case is taken from Studies in Expressive Movement , pp. 137-141, by per* 
mission of The Macmillan Company, 


3 58 

At the same time it explains almost everything else in his nature, serv- 
ing to reconcile other apparently discordant habits, traits, and atti- 
tudes. For such a cardinal trait, Wertheimer has proposed the desig- 
nation radix. The radix may never be sought on the activity level, for 
it lies at the root of activities. And however inconsistent activities 
may seem to be, they are congruent so long as they spring from the 

Separate orderly and disorderly 
acts (illusion of specificity) 

Organized but opposed expres* 
sive traits 

Cardinal trait or radix 
(D’s particular pattern of self- 

Figure 25 

An Illustration of Congruence 

(The unity of personality becomes apparent as more basic dynamic systems are 

same root. The argument as applied to the case of D may be illus- 
trated with the schematic diagram in Figure 25. 

The cardinal trait, or radix, has been represented in a slightly 
different way by H. A. Murray under the title of unity-thema. Ac- 
cording to this conception, too, the dominant tendencies of a life are 
said to derive from a single central dynamic principle, which if 
properly understood would explain both the collaborating and the 
conflicting actions of the person. But according to Murray, the unity- 
thema often springs from fixations formed early in life, and can be 
accurately discovered only through psychoanalytic exploration. 
Here, then, is the principle of congruence with a Freudian render- 

Yet the conception of congruence must not be carried too far. 


To say that whatever occurs in a personal life is ipso facto consistent 
with everything else is going to a pointless extreme. The following 
statement is an example of such exaggeration: 

The unity of a person can be traced in each instant of his life. 
There is nothing in character that contradicts itself. If a person who 
is known to us seems to be incongruous with himself that is only an 
indication of the inadequacy and superficiality of our previous obser- 
vations . 14 

The fallacy of this statement is its confusion of mere spacio-temporal 
unity (locality) with psychological unity. 

It is difficult to keep a balanced footing. On the one hand there 
is no doubt that most experimentalists have under-estimated the con- 
sistencies of personality simply because their methods for determin- 
ing consistency lack penetration. On the other hand, it is easy to 
read into discordant behavior some mythical unity, thereby mis- 
taking an arbitrary interpretation for fact. In the case of the artist 
discussed above, there is no certain proof that his differential ex- 
pressive behavior (his varying pressures) did in fact spring from one 
common root (his artistic style of work). To say so seems plausible 
enough, but in this particular case the congruence claimed has not 
actually been proved. 

But are there any tests of congruence? Can it ever be objectively 
established? The strict positivist answers negatively. Congruence, he 
says, is always a matter of the investigator’s interpretation; meaning- 
ful unity is impossible to demonstrate objectively. He complains be- 
cause a psychoanalyst, for example, finds that the root of a certain 
life is an unquenchable, if unconscious, hatred for the father. How 
does the analyst know? Only because he “feels” that all the informa- 
tion he has obtained through the analysis “clicks” into place with this 
basic “integrational hypothesis.” But might not another investigator 
feel the same satisfying click when he hits upon an entirely different 

And how is it with the case study, the most available of all the 
methods of representing congruence? Dogmatic interpretation seems 
to be a danger whatever the variety we employ, whether biography, 
autobiography, the clinical record, the life-history, psychiatric sum- 
mary, or the thumb-nail sketch. It is true that in practice the validity 

14 R. Franlte, “Gang und Character” (ed. by H. Bogen and O. Lipmarai), 
Beihefte, Zscb. f. ang. Psych oL, 1931, No. 58, p. 45. 



of case studies is seldom checked. The assumption is that, given an 
accumulation of incident, the one true pattern of unity will emerge 
by sheer virtue of the “systematic relevance” of one incident to 
another. Thus the case study is tested only by its internal intelligi- 
bility, by its self-consistency. Such a test has been called the “logico- 
meaningfui” criterion of integration . 15 Though unsatisfactory to a 
positivist, to many investigators it appeals as ultimately the soundest 

Deferring to Chapter XIX more detailed discussion of this prob- 
lem, we shall merely suggest here that the picture is not quite so 
dark as the positivist paints it. A pronouncement regarding the con- 
gruence of a life need not be merely an ipse dixit of the investigator, 
for which there is no possibility of proof or disproof. For one thing, 
there may be many investigators making the diagnosis. Should they 
all come to the same decision regarding congruence, the basic scien- 
tific requirement of agreement would be met. 

In such a multiple diagnosis, it is, of course, scientifically desirable 
to have the investigators work in strict independence of one another. 
But because of the deficiency in the skill of the various investigators, 
to achieve agreement is hard. Perhaps the majority opinion may be 
incorrect, and the minority correct. Conceivably, too, only one judge 
may be right in his interpretations. To offset this danger, conferences 
between the investigators may be arranged, wherein they pool their 
evidence and try to reach a synthetic interpretation. The conference 
procedure sacrifices the strict check of independent diagnosis, but it 
gains something from the corrective influence of consultation. The 
best variety of the method might be one that permits consultation on 
evidence but secures independence of diagnosis . 16 

It is a problem for the future to perfect methods for establishing 
congruence in personality. There is need for criteria to determine 
whether the pattern that is held to be congruent is really so, or 
whether it is being “read into” the case. Some progress has been made. 
Both prediction and matching have a certain usefulness here. So too 
does the method of multiple diagnosis discussed above. More and 

15 P- A. Sorokin, Rural SocioL, 1936, 1, 121-141, 344-374. The logico-meaningful 
method he defines as the determination of “the identity (or similarity) of central 
meaning, idea, or mental bias, that permeates all die logically related fragments” 
<P- 347)- 

16 The method of multiple diagnosis is illustrated very well in the forthcoming 
experimental study Explorations in Personality by H. A. Murray, et al. This study 
attempts to preserve the benefits of consultation between judges along with certain 
controls making for independence of diagnosis. 


more, psychiatrists and psychoanalysts are subjecting their diagnosis 
to checks. Even the case study will no doubt be improved and criteria 
evolved to determine its validity. This area of research is, without 
doubt, one of those most deserving of further exploration and 
methodological invention. 


Any school of psychological thought that aims deliberately to 
deal with ‘wholes is bound to say something about the unity of per- 
sonality. It is not surprising, therefore, that within recent years 
Gestalt psychology has made many contributions to this subject. 
Some of these (especially the contributions of Wertheimer, Koffka, 
and Kohler) are more appropriately discussed in other contexts, for 
they bear not so much on the problem of the unity of the total per- 
sonality as on the unity of its sub-structures, on its patterns of ex- 
pressive movement, and on the process of apprehending personality 
as an integral whole. 

The question for the moment is whether the total personality 
may be considered objectively as a Gestalt: a Gestalt being defined 
as a “system whose parts are dynamically connected in such a way 
that a change of one part results in a change of all other parts.” 17 
To this question Lewin gives an affirmative answer, and justifies his 
position by a large number of experimental findings. Conceptually 
he represents his theory of the structure of the personal Gestalt with 
“topological” diagrams. One of these is reproduced in Figure 26. 13 

The person is regarded as a differentiated region separated by a 
permeable boundary from his external environment. In direct con- 
tact with the environment are certain perceptual-motor systems en- 
gaged in sensing and in acting adaptively. One of the most important 
processes of this region, for example, is speech which plays a princi- 
pal role in the communication between the person and his social 
environment. As a rule the perceptual-motor region possesses a rela- 
tively high unity; its systems are fluid in that they readily interact, 
and the region as a whole is often called into play when adaptive 
acts are demanded. 

Inward from this perceptual-motor region lie the peripheral 

17 K. Lewin, Principles of Topological Psychology , 1936, p. 218. 

18 Reproduced by permission from Principles of Topological Psychology 
(McGraw-Hill Book Co.), p. 177. 


regions of the inner personality. These are more enduring and ordi- 
narily better structured and more clearly segregated from one 
another than are the motor-perceptual regions. In some respects these 
peripheral regions seem to correspond to the directive or expressive 
traits discussed on pp. 323 f. When under tension they are re- 
garded as systems, that may act either independently or— more likely 

Topological Representation of the Structure of Personality 

(My motor-perceptual region; /, inner-personal region; P, peripheral parts of 
C, central parts of I; E, environment.) 

—as parts of larger systems. It is a conception similar to the account 
given on p. 326 of the activation of interdependent traits. 

Proceeding inward we find more central regions, representing, it 
seems, the principal driving traits of personality, the deeper motives 
and interests, and more lasting sentiments and prejudices. These cen- 
tral regions are not unlike the “central” traits described on pp. 
337 f. Finally, at the heart of the inner-personal region lies the very 
most intimate zone of all, elsewhere called by Lewin the “core.” It 
is this intimate center that is aroused in states of acute self-conscious 
emotion and is always involved in aspiration and in fantasy. It is the 
region that is most difficult for another to penetrate or to understand. 
Yet it is a region whose existence is attested not only by introspection 
but by a number of experimental researches. 10 This core, more than 

19 K. Lewin, Principles of T opological Psychology, 1936, p. 180. 


any other single factor, guarantees the relative stability of the total 
structure. Though in some respects unique, Lewin’s conception of 
the core of personality is not unrelated to the older principle of 

The sub-structures of personality here described are not to be 
regarded as rigid. Their boundaries are relatively permeable, and the 
whole organization is to some degree fluid, as shown by several facts. 
For one thing, any intense stimulation is more than likely to affect 
the state of the whole person, easily dissipating the otherwise effec- 
tive functional barriers. Further, though the temporal dimension is 
not represented in topology, it is admitted that with time the regions 
alter greatly in their contours and scope. With time they tend to 
become finer and more segregated (the principle of differentiation), 
and tend likewise to restructure themselves in different patterns. 
“Young” systems may become stronger than the older systems, and 
may dynamically be quite independent of them (principle of func- 
tional autonomy). In regression, it is clear, there may be a return 
to older systems and a surrender of the newer. 20 

We have then a picture of personality as a stratified system, 
having a definite structure with distinguishable separate regions; but 
at the same time allowing for any degree of fusion and interde- 
pendence of the parts, for any amount of change— progressive or 
regressive— and for all kinds of individual differences. Personality, 
Lewin says in effect, is indeed a Gestalt, but it is a Gestalt that has 
greater or less unity, depending upon its own individual nature, upon 
the condition of the organism and the field in which it is behaving. 
Perhaps no other psychologist has succeeded quite so well in depict- 
ing at one and the same time so many of the intricate issues involved, 
and in demonstrating that the unity of personality is always a matter 
of degree. 

An additional virtue of the topological method is its ability to 
represent the individual case, with all its peculiarities. Indeed it is 
one of the most insistent demands of topological psychology that the 
angle case be accorded a position of highest respectability in psycho- 
logical research. Lewin goes so far as to insist that scientific predic- 
tions can never successfully be made by a knowledge of general laws 
alone, derived as they are from abstractive analysis. The topology 

20 Ibid., p. 190. 



of the person in all its individual peculiarity must be understood 
before its true dynamic tendencies can be discerned. 21 

A further extension of the topological method adds a third dL 
mension to the diagram, an “irreality level.” By this means Lewin 
seeks to represent that in fantasy, in delusions, or in play, the struc- 
ture of personality is somewhat altered. Whether this addition is of 
any great advantage may be questioned, for in all the regions of the 
personal life there are many aspects that might be called irreal. Yet 
the real and the irreal are bound into well-knit integral dispositions 
or systems. Ultimately it is impossible to know what in any psycho- 
logical state is real and what irreal. For what, pray tell, is reality? 
Is an ambition real or irreal? and what about an ego-ideal or a guid- 
ing image? 

The principal advantage generally claimed for the topological 
treatment of personality has not as yet been mentioned. It brings 
the person within the domain of the so-called “field theory,” which 
endeavors to tre